[{"quality_controlled":"1","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2206.05181"}],"page":"2128-2134","title":"Lightweight conditional model extrapolation for streaming data under class-prior shift","oa_version":"Preprint","publication_status":"published","abstract":[{"lang":"eng","text":"We introduce LIMES, a new method for learning with non-stationary streaming data, inspired by the recent success of meta-learning. The main idea is not to attempt to learn a single classifier that would have to work well across all occurring data distributions, nor many separate classifiers, but to exploit a hybrid strategy: we learn a single set of model parameters from which a specific classifier for any specific data distribution is derived via classifier adaptation. Assuming a multiclass classification setting with class-prior shift, the adaptation step can be performed analytically with only the classifier’s bias terms being affected. Another contribution of our work is an extrapolation step that predicts suitable adaptation parameters for future time steps based on the previous data. In combination, we obtain a lightweight procedure for learning from streaming data with varying class distribution that adds no trainable parameters and almost no memory or computational overhead compared to training a single model. Experiments on a set of exemplary tasks using Twitter data show that LIMES achieves higher accuracy than alternative approaches, especially with respect to the relevant real-world metric of lowest within-day accuracy."}],"user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","citation":{"ieee":"P. Tomaszewska and C. Lampert, “Lightweight conditional model extrapolation for streaming data under class-prior shift,” in <i>26th International Conference on Pattern Recognition</i>, Montreal, Canada, 2022, vol. 2022, pp. 2128–2134.","ista":"Tomaszewska P, Lampert C. 2022. Lightweight conditional model extrapolation for streaming data under class-prior shift. 26th International Conference on Pattern Recognition. ICPR: International Conference on Pattern Recognition vol. 2022, 2128–2134.","apa":"Tomaszewska, P., &#38; Lampert, C. (2022). Lightweight conditional model extrapolation for streaming data under class-prior shift. In <i>26th International Conference on Pattern Recognition</i> (Vol. 2022, pp. 2128–2134). Montreal, Canada: Institute of Electrical and Electronics Engineers. <a href=\"https://doi.org/10.1109/icpr56361.2022.9956195\">https://doi.org/10.1109/icpr56361.2022.9956195</a>","chicago":"Tomaszewska, Paulina, and Christoph Lampert. “Lightweight Conditional Model Extrapolation for Streaming Data under Class-Prior Shift.” In <i>26th International Conference on Pattern Recognition</i>, 2022:2128–34. Institute of Electrical and Electronics Engineers, 2022. <a href=\"https://doi.org/10.1109/icpr56361.2022.9956195\">https://doi.org/10.1109/icpr56361.2022.9956195</a>.","mla":"Tomaszewska, Paulina, and Christoph Lampert. “Lightweight Conditional Model Extrapolation for Streaming Data under Class-Prior Shift.” <i>26th International Conference on Pattern Recognition</i>, vol. 2022, Institute of Electrical and Electronics Engineers, 2022, pp. 2128–34, doi:<a href=\"https://doi.org/10.1109/icpr56361.2022.9956195\">10.1109/icpr56361.2022.9956195</a>.","short":"P. Tomaszewska, C. Lampert, in:, 26th International Conference on Pattern Recognition, Institute of Electrical and Electronics Engineers, 2022, pp. 2128–2134.","ama":"Tomaszewska P, Lampert C. Lightweight conditional model extrapolation for streaming data under class-prior shift. In: <i>26th International Conference on Pattern Recognition</i>. Vol 2022. Institute of Electrical and Electronics Engineers; 2022:2128-2134. doi:<a href=\"https://doi.org/10.1109/icpr56361.2022.9956195\">10.1109/icpr56361.2022.9956195</a>"},"arxiv":1,"date_published":"2022-11-29T00:00:00Z","oa":1,"scopus_import":"1","_id":"12161","author":[{"full_name":"Tomaszewska, Paulina","first_name":"Paulina","last_name":"Tomaszewska"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph","last_name":"Lampert","orcid":"0000-0001-8622-7887","first_name":"Christoph"}],"doi":"10.1109/icpr56361.2022.9956195","date_created":"2023-01-12T12:09:38Z","volume":2022,"publication_identifier":{"eissn":["2831-7475"],"eisbn":["9781665490627"]},"intvolume":"      2022","department":[{"_id":"ChLa"}],"isi":1,"publication":"26th International Conference on Pattern Recognition","article_processing_charge":"No","status":"public","external_id":{"isi":["000897707602018"],"arxiv":["2206.05181"]},"date_updated":"2024-10-09T21:03:41Z","language":[{"iso":"eng"}],"day":"29","month":"11","corr_author":"1","publisher":"Institute of Electrical and Electronics Engineers","conference":{"end_date":"2022-08-25","name":"ICPR: International Conference on Pattern Recognition","start_date":"2022-08-21","location":"Montreal, Canada"},"year":"2022","type":"conference"},{"article_type":"original","publication_identifier":{"issn":["2835-8856"]},"acknowledgement":"The authors would like to thank Bernd Prach, Elias Frantar, Alexandra Peste, Mahdi Nikdan, and Peter Súkeník for their helpful feedback. This research was supported by the Scientific Service Units (SSU) of IST Austria through resources provided by Scientific Computing (SciComp). This publication was made possible by an ETH AI Center postdoctoral fellowship granted to Nikola Konstantinov. Eugenia Iofinova was supported in part by the FWF DK VGSCO, grant agreement number W1260-N35. ","arxiv":1,"citation":{"ista":"Iofinova EB, Konstantinov NH, Lampert C. 2022. FLEA: Provably robust fair multisource learning from unreliable training data. Transactions on Machine Learning Research.","ieee":"E. B. Iofinova, N. H. Konstantinov, and C. Lampert, “FLEA: Provably robust fair multisource learning from unreliable training data,” <i>Transactions on Machine Learning Research</i>. ML Research Press, 2022.","ama":"Iofinova EB, Konstantinov NH, Lampert C. FLEA: Provably robust fair multisource learning from unreliable training data. <i>Transactions on Machine Learning Research</i>. 2022.","mla":"Iofinova, Eugenia B., et al. “FLEA: Provably Robust Fair Multisource Learning from Unreliable Training Data.” <i>Transactions on Machine Learning Research</i>, ML Research Press, 2022.","short":"E.B. Iofinova, N.H. Konstantinov, C. Lampert, Transactions on Machine Learning Research (2022).","chicago":"Iofinova, Eugenia B, Nikola H Konstantinov, and Christoph Lampert. “FLEA: Provably Robust Fair Multisource Learning from Unreliable Training Data.” <i>Transactions on Machine Learning Research</i>. ML Research Press, 2022.","apa":"Iofinova, E. B., Konstantinov, N. H., &#38; Lampert, C. (2022). FLEA: Provably robust fair multisource learning from unreliable training data. <i>Transactions on Machine Learning Research</i>. ML Research Press."},"date_published":"2022-12-22T00:00:00Z","oa":1,"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","author":[{"first_name":"Eugenia B","orcid":"0000-0002-7778-3221","last_name":"Iofinova","full_name":"Iofinova, Eugenia B","id":"f9a17499-f6e0-11ea-865d-fdf9a3f77117"},{"last_name":"Konstantinov","orcid":"0009-0009-5204-7621","first_name":"Nikola H","id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87","full_name":"Konstantinov, Nikola H"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887","last_name":"Lampert","first_name":"Christoph"}],"date_created":"2023-02-02T20:29:57Z","_id":"12495","publication_status":"published","title":"FLEA: Provably robust fair multisource learning from unreliable training data","oa_version":"Published Version","acknowledged_ssus":[{"_id":"ScienComp"}],"abstract":[{"lang":"eng","text":"Fairness-aware learning aims at constructing classifiers that not only make accurate predictions, but also do not discriminate against specific groups. It is a fast-growing area of\r\nmachine learning with far-reaching societal impact. However, existing fair learning methods\r\nare vulnerable to accidental or malicious artifacts in the training data, which can cause\r\nthem to unknowingly produce unfair classifiers. In this work we address the problem of\r\nfair learning from unreliable training data in the robust multisource setting, where the\r\navailable training data comes from multiple sources, a fraction of which might not be representative of the true data distribution. We introduce FLEA, a filtering-based algorithm\r\nthat identifies and suppresses those data sources that would have a negative impact on\r\nfairness or accuracy if they were used for training. As such, FLEA is not a replacement of\r\nprior fairness-aware learning methods but rather an augmentation that makes any of them\r\nrobust against unreliable training data. We show the effectiveness of our approach by a\r\ndiverse range of experiments on multiple datasets. Additionally, we prove formally that\r\n–given enough data– FLEA protects the learner against corruptions as long as the fraction of\r\naffected data sources is less than half. Our source code and documentation are available at\r\nhttps://github.com/ISTAustria-CVML/FLEA."}],"file_date_updated":"2023-02-23T10:30:04Z","quality_controlled":"1","main_file_link":[{"open_access":"1","url":"https://openreview.net/forum?id=XsPopigZXV"}],"type":"journal_article","project":[{"_id":"9B9290DE-BA93-11EA-9121-9846C619BF3A","name":"Vienna Graduate School on Computational Optimization","grant_number":"W1260-N35"}],"year":"2022","tmp":{"name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","short":"CC BY (4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","image":"/images/cc_by.png"},"day":"22","publisher":"ML Research Press","month":"12","corr_author":"1","ddc":["000"],"date_updated":"2025-12-30T11:04:31Z","file":[{"relation":"main_file","checksum":"97c8a8470759cab597abb973ca137a3b","file_name":"2022_TMLR_Iofinova.pdf","file_id":"12673","date_created":"2023-02-23T10:30:04Z","creator":"dernst","access_level":"open_access","content_type":"application/pdf","date_updated":"2023-02-23T10:30:04Z","file_size":1948063,"success":1}],"language":[{"iso":"eng"}],"related_material":{"link":[{"url":"https://github.com/ISTAustria-CVML/FLEA","relation":"software","description":"source code"}]},"article_processing_charge":"No","has_accepted_license":"1","department":[{"_id":"ChLa"}],"publication":"Transactions on Machine Learning Research","external_id":{"arxiv":["2106.11732"]},"status":"public"},{"quality_controlled":"1","main_file_link":[{"url":"https://arxiv.org/abs/2102.06004","open_access":"1"}],"page":"59-83","title":"On the impossibility of fairness-aware learning from corrupted data","oa_version":"Preprint","publication_status":"published","abstract":[{"lang":"eng","text":"Addressing fairness concerns about machine learning models is a crucial step towards their long-term adoption in real-world automated systems. Many approaches for training fair models from data have been developed and an implicit assumption about such algorithms is that they are able to recover a fair model, despite potential historical biases in the data. In this work we show a number of impossibility results that indicate that there is no learning algorithm that can recover a fair model when a proportion of the dataset is subject to arbitrary manipulations. Specifically, we prove that there are situations in which an adversary can force any learner to return a biased classifier, with or without degrading accuracy, and that the strength of this bias increases for learning problems with underrepresented protected groups in the data. Our results emphasize on the importance of studying further data corruption models of various strength and of establishing stricter data collection practices for fairness-aware learning."}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","arxiv":1,"citation":{"ista":"Konstantinov NH, Lampert C. 2022. On the impossibility of fairness-aware learning from corrupted data. Proceedings of Machine Learning Research. vol. 171, 59–83.","ieee":"N. H. Konstantinov and C. Lampert, “On the impossibility of fairness-aware learning from corrupted data,” in <i>Proceedings of Machine Learning Research</i>, 2022, vol. 171, pp. 59–83.","ama":"Konstantinov NH, Lampert C. On the impossibility of fairness-aware learning from corrupted data. In: <i>Proceedings of Machine Learning Research</i>. Vol 171. ML Research Press; 2022:59-83.","mla":"Konstantinov, Nikola H., and Christoph Lampert. “On the Impossibility of Fairness-Aware Learning from Corrupted Data.” <i>Proceedings of Machine Learning Research</i>, vol. 171, ML Research Press, 2022, pp. 59–83.","short":"N.H. Konstantinov, C. Lampert, in:, Proceedings of Machine Learning Research, ML Research Press, 2022, pp. 59–83.","chicago":"Konstantinov, Nikola H, and Christoph Lampert. “On the Impossibility of Fairness-Aware Learning from Corrupted Data.” In <i>Proceedings of Machine Learning Research</i>, 171:59–83. ML Research Press, 2022.","apa":"Konstantinov, N. H., &#38; Lampert, C. (2022). On the impossibility of fairness-aware learning from corrupted data. In <i>Proceedings of Machine Learning Research</i> (Vol. 171, pp. 59–83). ML Research Press."},"date_published":"2022-12-01T00:00:00Z","oa":1,"scopus_import":"1","_id":"13241","author":[{"last_name":"Konstantinov","first_name":"Nikola H","id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87","full_name":"Konstantinov, Nikola H"},{"first_name":"Christoph","orcid":"0000-0001-8622-7887","last_name":"Lampert","full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"date_created":"2023-07-16T22:01:13Z","volume":171,"publication_identifier":{"eissn":["2640-3498"]},"acknowledgement":"This paper is a shortened, workshop version of Konstantinov and Lampert (2021),\r\nhttps://arxiv.org/abs/2102.06004. For further results, including an analysis of algorithms achieving the lower bounds from this paper, we refer to the full version.","intvolume":"       171","department":[{"_id":"ChLa"}],"publication":"Proceedings of Machine Learning Research","article_processing_charge":"No","status":"public","external_id":{"arxiv":["2102.06004"]},"date_updated":"2024-10-09T21:05:54Z","language":[{"iso":"eng"}],"related_material":{"record":[{"id":"10802","status":"public","relation":"extended_version"}]},"day":"01","month":"12","corr_author":"1","publisher":"ML Research Press","year":"2022","type":"conference"},{"quality_controlled":"1","page":"5185-5192","publication_status":"published","title":"Overcoming rare-language discrimination in multi-lingual sentiment analysis","oa_version":"None","acknowledged_ssus":[{"_id":"ScienComp"}],"abstract":[{"text":"The digitalization of almost all aspects of our everyday lives has led to unprecedented amounts of data being freely available on the Internet. In particular social media platforms provide rich sources of user-generated data, though typically in unstructured form, and with high diversity, such as written in many different languages. Automatically identifying meaningful information in such big data resources and extracting it efficiently is one of the ongoing challenges of our time. A common step for this is sentiment analysis, which forms the foundation for tasks such as opinion mining or trend prediction. Unfortunately, publicly available tools for this task are almost exclusively available for English-language texts. Consequently, a large fraction of the Internet users, who do not communicate in English, are ignored in automatized studies, a phenomenon called rare-language discrimination.In this work we propose a technique to overcome this problem by a truly multi-lingual model, which can be trained automatically without linguistic knowledge or even the ability to read the many target languages. The main step is to combine self-annotation, specifically the use of emoticons as a proxy for labels, with multi-lingual sentence representations.To evaluate our method we curated several large datasets from data obtained via the free Twitter streaming API. The results show that our proposed multi-lingual training is able to achieve sentiment predictions at the same quality level for rare languages as for frequent ones, and in particular clearly better than what mono-lingual training achieves on the same data. ","lang":"eng"}],"citation":{"ista":"Lampert J, Lampert C. 2022. Overcoming rare-language discrimination in multi-lingual sentiment analysis. 2021 IEEE International Conference on Big Data. Big Data: International Conference on Big Data, 5185–5192.","ieee":"J. Lampert and C. Lampert, “Overcoming rare-language discrimination in multi-lingual sentiment analysis,” in <i>2021 IEEE International Conference on Big Data</i>, Orlando, FL, United States, 2022, pp. 5185–5192.","mla":"Lampert, Jasmin, and Christoph Lampert. “Overcoming Rare-Language Discrimination in Multi-Lingual Sentiment Analysis.” <i>2021 IEEE International Conference on Big Data</i>, IEEE, 2022, pp. 5185–92, doi:<a href=\"https://doi.org/10.1109/bigdata52589.2021.9672003\">10.1109/bigdata52589.2021.9672003</a>.","short":"J. Lampert, C. Lampert, in:, 2021 IEEE International Conference on Big Data, IEEE, 2022, pp. 5185–5192.","chicago":"Lampert, Jasmin, and Christoph Lampert. “Overcoming Rare-Language Discrimination in Multi-Lingual Sentiment Analysis.” In <i>2021 IEEE International Conference on Big Data</i>, 5185–92. IEEE, 2022. <a href=\"https://doi.org/10.1109/bigdata52589.2021.9672003\">https://doi.org/10.1109/bigdata52589.2021.9672003</a>.","ama":"Lampert J, Lampert C. Overcoming rare-language discrimination in multi-lingual sentiment analysis. In: <i>2021 IEEE International Conference on Big Data</i>. IEEE; 2022:5185-5192. doi:<a href=\"https://doi.org/10.1109/bigdata52589.2021.9672003\">10.1109/bigdata52589.2021.9672003</a>","apa":"Lampert, J., &#38; Lampert, C. (2022). Overcoming rare-language discrimination in multi-lingual sentiment analysis. In <i>2021 IEEE International Conference on Big Data</i> (pp. 5185–5192). Orlando, FL, United States: IEEE. <a href=\"https://doi.org/10.1109/bigdata52589.2021.9672003\">https://doi.org/10.1109/bigdata52589.2021.9672003</a>"},"date_published":"2022-01-13T00:00:00Z","user_id":"317138e5-6ab7-11ef-aa6d-ffef3953e345","doi":"10.1109/bigdata52589.2021.9672003","date_created":"2022-02-10T14:08:23Z","author":[{"full_name":"Lampert, Jasmin","first_name":"Jasmin","last_name":"Lampert"},{"first_name":"Christoph","last_name":"Lampert","orcid":"0000-0002-4561-241X","full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"scopus_import":"1","_id":"10752","publication_identifier":{"isbn":["9781665439022"]},"acknowledgement":"This research was funded in parts by the FORTE program of the Austrian Research Promotion Agency (FFG) and the Federal Ministry of Agriculture, Regions and Tourism (BMLRT) as part of the AMMONIS project (grant no. 879705). The research was also supported by the Scientific Service Units (SSU) of IST Austria through resources provided by Scientific Computing (SciComp).","article_processing_charge":"No","department":[{"_id":"ChLa"}],"isi":1,"publication":"2021 IEEE International Conference on Big Data","external_id":{"isi":["000800559505036"]},"status":"public","date_updated":"2024-10-21T06:01:53Z","language":[{"iso":"eng"}],"day":"13","publisher":"IEEE","conference":{"location":"Orlando, FL, United States","start_date":"2021-12-15","name":"Big Data: International Conference on Big Data","end_date":"2021-12-18"},"month":"01","corr_author":"1","type":"conference","year":"2022"},{"status":"public","external_id":{"arxiv":["2208.03160"],"isi":["000904104000021"]},"publication":"Computer Vision – ECCV 2022","department":[{"_id":"GradSch"},{"_id":"ChLa"}],"isi":1,"article_processing_charge":"No","related_material":{"record":[{"relation":"dissertation_contains","id":"19759","status":"public"}]},"language":[{"iso":"eng"}],"date_updated":"2026-04-07T11:49:51Z","corr_author":"1","month":"10","conference":{"end_date":"2022-10-27","start_date":"2022-10-23","name":"ECCV: European Conference on Computer Vision","location":"Tel Aviv, Israel"},"publisher":"Springer Nature","day":"23","year":"2022","type":"conference","main_file_link":[{"url":" https://doi.org/10.48550/arXiv.2208.03160","open_access":"1"}],"page":"350-365","quality_controlled":"1","abstract":[{"text":"It is a highly desirable property for deep networks to be robust against\r\nsmall input changes. One popular way to achieve this property is by designing\r\nnetworks with a small Lipschitz constant. In this work, we propose a new\r\ntechnique for constructing such Lipschitz networks that has a number of\r\ndesirable properties: it can be applied to any linear network layer\r\n(fully-connected or convolutional), it provides formal guarantees on the\r\nLipschitz constant, it is easy to implement and efficient to run, and it can be\r\ncombined with any training objective and optimization method. In fact, our\r\ntechnique is the first one in the literature that achieves all of these\r\nproperties simultaneously. Our main contribution is a rescaling-based weight\r\nmatrix parametrization that guarantees each network layer to have a Lipschitz\r\nconstant of at most 1 and results in the learned weight matrices to be close to\r\northogonal. Hence we call such layers almost-orthogonal Lipschitz (AOL).\r\nExperiments and ablation studies in the context of image classification with\r\ncertified robust accuracy confirm that AOL layers achieve results that are on\r\npar with most existing methods. Yet, they are simpler to implement and more\r\nbroadly applicable, because they do not require computationally expensive\r\nmatrix orthogonalization or inversion steps as part of the network\r\narchitecture. We provide code at https://github.com/berndprach/AOL.","lang":"eng"}],"oa_version":"Preprint","title":"Almost-orthogonal layers for efficient general-purpose Lipschitz networks","publication_status":"published","_id":"11839","scopus_import":"1","doi":"10.1007/978-3-031-19803-8_21","date_created":"2022-08-12T15:09:47Z","author":[{"id":"2D561D42-C427-11E9-89B4-9C1AE6697425","full_name":"Prach, Bernd","last_name":"Prach","first_name":"Bernd"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph","last_name":"Lampert","orcid":"0000-0001-8622-7887","first_name":"Christoph"}],"user_id":"317138e5-6ab7-11ef-aa6d-ffef3953e345","date_published":"2022-10-23T00:00:00Z","oa":1,"arxiv":1,"citation":{"ieee":"B. Prach and C. Lampert, “Almost-orthogonal layers for efficient general-purpose Lipschitz networks,” in <i>Computer Vision – ECCV 2022</i>, Tel Aviv, Israel, 2022, vol. 13681, pp. 350–365.","ista":"Prach B, Lampert C. 2022. Almost-orthogonal layers for efficient general-purpose Lipschitz networks. Computer Vision – ECCV 2022. ECCV: European Conference on Computer Vision, LNCS, vol. 13681, 350–365.","apa":"Prach, B., &#38; Lampert, C. (2022). Almost-orthogonal layers for efficient general-purpose Lipschitz networks. In <i>Computer Vision – ECCV 2022</i> (Vol. 13681, pp. 350–365). Tel Aviv, Israel: Springer Nature. <a href=\"https://doi.org/10.1007/978-3-031-19803-8_21\">https://doi.org/10.1007/978-3-031-19803-8_21</a>","ama":"Prach B, Lampert C. Almost-orthogonal layers for efficient general-purpose Lipschitz networks. In: <i>Computer Vision – ECCV 2022</i>. Vol 13681. Springer Nature; 2022:350-365. doi:<a href=\"https://doi.org/10.1007/978-3-031-19803-8_21\">10.1007/978-3-031-19803-8_21</a>","short":"B. Prach, C. Lampert, in:, Computer Vision – ECCV 2022, Springer Nature, 2022, pp. 350–365.","mla":"Prach, Bernd, and Christoph Lampert. “Almost-Orthogonal Layers for Efficient General-Purpose Lipschitz Networks.” <i>Computer Vision – ECCV 2022</i>, vol. 13681, Springer Nature, 2022, pp. 350–65, doi:<a href=\"https://doi.org/10.1007/978-3-031-19803-8_21\">10.1007/978-3-031-19803-8_21</a>.","chicago":"Prach, Bernd, and Christoph Lampert. “Almost-Orthogonal Layers for Efficient General-Purpose Lipschitz Networks.” In <i>Computer Vision – ECCV 2022</i>, 13681:350–65. Springer Nature, 2022. <a href=\"https://doi.org/10.1007/978-3-031-19803-8_21\">https://doi.org/10.1007/978-3-031-19803-8_21</a>."},"intvolume":"     13681","volume":13681,"publication_identifier":{"isbn":["9783031198021"],"eisbn":["9783031198038"]},"alternative_title":["LNCS"]},{"year":"2022","project":[{"name":"Vienna Graduate School on Computational Optimization","grant_number":"W1260-N35","_id":"9B9290DE-BA93-11EA-9121-9846C619BF3A"},{"_id":"268A44D6-B435-11E9-9278-68D0E5697425","grant_number":"805223","call_identifier":"H2020","name":"Elastic Coordination for Scalable Machine Learning"}],"type":"conference","ec_funded":1,"day":"27","month":"09","corr_author":"1","publisher":"Institute of Electrical and Electronics Engineers","conference":{"location":"New Orleans, LA, United States","name":"CVPR: Computer Vision and Pattern Recognition","start_date":"2022-06-18","end_date":"2022-06-24"},"date_updated":"2026-04-07T13:30:19Z","language":[{"iso":"eng"}],"related_material":{"record":[{"relation":"dissertation_contains","status":"public","id":"13074"}]},"isi":1,"department":[{"_id":"DaAl"},{"_id":"ChLa"}],"publication":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition","article_processing_charge":"No","status":"public","external_id":{"arxiv":["2111.13445"],"isi":["000870759105034"]},"publication_identifier":{"eissn":["2575-7075"]},"acknowledgement":"he authors would like to sincerely thank Christoph Lampert and Nir Shavit for fruitful discussions during the development of this work, and Eldar Kurtic for experimental support. EI was supported in part by the FWF DK VGSCO, grant agreement number W1260-N35, while AP and DA acknowledge generous support by the ERC, via Starting Grant 805223 ScaleML.","user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","citation":{"ieee":"E. B. Iofinova, A. Krumes, M. Kurtz, and D.-A. Alistarh, “How well do sparse ImageNet models transfer?,” in <i>2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>, New Orleans, LA, United States, 2022, pp. 12256–12266.","ista":"Iofinova EB, Krumes A, Kurtz M, Alistarh D-A. 2022. How well do sparse ImageNet models transfer? 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. CVPR: Computer Vision and Pattern Recognition, 12256–12266.","apa":"Iofinova, E. B., Krumes, A., Kurtz, M., &#38; Alistarh, D.-A. (2022). How well do sparse ImageNet models transfer? In <i>2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition</i> (pp. 12256–12266). New Orleans, LA, United States: Institute of Electrical and Electronics Engineers. <a href=\"https://doi.org/10.1109/cvpr52688.2022.01195\">https://doi.org/10.1109/cvpr52688.2022.01195</a>","ama":"Iofinova EB, Krumes A, Kurtz M, Alistarh D-A. How well do sparse ImageNet models transfer? In: <i>2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>. Institute of Electrical and Electronics Engineers; 2022:12256-12266. doi:<a href=\"https://doi.org/10.1109/cvpr52688.2022.01195\">10.1109/cvpr52688.2022.01195</a>","short":"E.B. Iofinova, A. Krumes, M. Kurtz, D.-A. Alistarh, in:, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Institute of Electrical and Electronics Engineers, 2022, pp. 12256–12266.","chicago":"Iofinova, Eugenia B, Alexandra Krumes, Mark Kurtz, and Dan-Adrian Alistarh. “How Well Do Sparse ImageNet Models Transfer?” In <i>2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>, 12256–66. Institute of Electrical and Electronics Engineers, 2022. <a href=\"https://doi.org/10.1109/cvpr52688.2022.01195\">https://doi.org/10.1109/cvpr52688.2022.01195</a>.","mla":"Iofinova, Eugenia B., et al. “How Well Do Sparse ImageNet Models Transfer?” <i>2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>, Institute of Electrical and Electronics Engineers, 2022, pp. 12256–66, doi:<a href=\"https://doi.org/10.1109/cvpr52688.2022.01195\">10.1109/cvpr52688.2022.01195</a>."},"arxiv":1,"oa":1,"date_published":"2022-09-27T00:00:00Z","scopus_import":"1","_id":"12299","date_created":"2023-01-16T10:06:00Z","doi":"10.1109/cvpr52688.2022.01195","author":[{"first_name":"Eugenia B","orcid":"0000-0002-7778-3221","last_name":"Iofinova","full_name":"Iofinova, Eugenia B","id":"f9a17499-f6e0-11ea-865d-fdf9a3f77117"},{"full_name":"Peste, Elena-Alexandra","id":"32D78294-F248-11E8-B48F-1D18A9856A87","first_name":"Elena-Alexandra","last_name":"Peste"},{"full_name":"Kurtz, Mark","last_name":"Kurtz","first_name":"Mark"},{"full_name":"Alistarh, Dan-Adrian","id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","first_name":"Dan-Adrian","orcid":"0000-0003-3650-940X","last_name":"Alistarh"}],"title":"How well do sparse ImageNet models transfer?","oa_version":"Preprint","publication_status":"published","abstract":[{"lang":"eng","text":"Transfer learning is a classic paradigm by which models pretrained on large “upstream” datasets are adapted to yield good results on “downstream” specialized datasets. Generally, more accurate models on the “upstream” dataset tend to provide better transfer accuracy “downstream”. In this work, we perform an in-depth investigation of this phenomenon in the context of convolutional neural networks (CNNs) trained on the ImageNet dataset, which have been pruned-that is, compressed by sparsifiying their connections. We consider transfer using unstructured pruned models obtained by applying several state-of-the-art pruning methods, including magnitude-based, second-order, regrowth, lottery-ticket, and regularization approaches, in the context of twelve standard transfer tasks. In a nutshell, our study shows that sparse models can match or even outperform the transfer performance of dense models, even at high sparsities, and, while doing so, can lead to significant inference and even training speedups. At the same time, we observe and analyze significant differences in the behaviour of different pruning methods. The code is available at: https://github.com/IST-DASLab/sparse-imagenet-transfer."}],"quality_controlled":"1","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2111.13445"}],"page":"12256-12266"},{"page":"176","file_date_updated":"2022-03-10T12:11:48Z","abstract":[{"text":"Because of the increasing popularity of machine learning methods, it is becoming important to understand the impact of learned components on automated decision-making systems and to guarantee that their consequences are beneficial to society. In other words, it is necessary to ensure that machine learning is sufficiently trustworthy to be used in real-world applications. This thesis studies two properties of machine learning models that are highly desirable for the\r\nsake of reliability: robustness and fairness. In the first part of the thesis we study the robustness of learning algorithms to training data corruption. Previous work has shown that machine learning models are vulnerable to a range\r\nof training set issues, varying from label noise through systematic biases to worst-case data manipulations. This is an especially relevant problem from a present perspective, since modern machine learning methods are particularly data hungry and therefore practitioners often have to rely on data collected from various external sources, e.g. from the Internet, from app users or via crowdsourcing. Naturally, such sources vary greatly in the quality and reliability of the\r\ndata they provide. With these considerations in mind, we study the problem of designing machine learning algorithms that are robust to corruptions in data coming from multiple sources. We show that, in contrast to the case of a single dataset with outliers, successful learning within this model is possible both theoretically and practically, even under worst-case data corruptions. The second part of this thesis deals with fairness-aware machine learning. There are multiple areas where machine learning models have shown promising results, but where careful considerations are required, in order to avoid discrimanative decisions taken by such learned components. Ensuring fairness can be particularly challenging, because real-world training datasets are expected to contain various forms of historical bias that may affect the learning process. In this thesis we show that data corruption can indeed render the problem of achieving fairness impossible, by tightly characterizing the theoretical limits of fair learning under worst-case data manipulations. However, assuming access to clean data, we also show how fairness-aware learning can be made practical in contexts beyond binary classification, in particular in the challenging learning to rank setting.","lang":"eng"}],"oa_version":"Published Version","title":"Robustness and fairness in machine learning","publication_status":"published","_id":"10799","OA_place":"publisher","date_created":"2022-02-28T13:03:49Z","author":[{"id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87","full_name":"Konstantinov, Nikola H","last_name":"Konstantinov","orcid":"0009-0009-5204-7621","first_name":"Nikola H"}],"doi":"10.15479/at:ista:10799","user_id":"ba8df636-2132-11f1-aed0-ed93e2281fdd","oa":1,"date_published":"2022-03-08T00:00:00Z","citation":{"mla":"Konstantinov, Nikola H. <i>Robustness and Fairness in Machine Learning</i>. Institute of Science and Technology Austria, 2022, doi:<a href=\"https://doi.org/10.15479/at:ista:10799\">10.15479/at:ista:10799</a>.","short":"N.H. Konstantinov, Robustness and Fairness in Machine Learning, Institute of Science and Technology Austria, 2022.","chicago":"Konstantinov, Nikola H. “Robustness and Fairness in Machine Learning.” Institute of Science and Technology Austria, 2022. <a href=\"https://doi.org/10.15479/at:ista:10799\">https://doi.org/10.15479/at:ista:10799</a>.","ama":"Konstantinov NH. Robustness and fairness in machine learning. 2022. doi:<a href=\"https://doi.org/10.15479/at:ista:10799\">10.15479/at:ista:10799</a>","apa":"Konstantinov, N. H. (2022). <i>Robustness and fairness in machine learning</i>. Institute of Science and Technology Austria. <a href=\"https://doi.org/10.15479/at:ista:10799\">https://doi.org/10.15479/at:ista:10799</a>","ista":"Konstantinov NH. 2022. Robustness and fairness in machine learning. Institute of Science and Technology Austria.","ieee":"N. H. Konstantinov, “Robustness and fairness in machine learning,” Institute of Science and Technology Austria, 2022."},"supervisor":[{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","first_name":"Christoph","orcid":"0000-0001-8622-7887","last_name":"Lampert"}],"publication_identifier":{"issn":["2663-337X"],"isbn":["978-3-99078-015-2"]},"alternative_title":["ISTA Thesis"],"status":"public","has_accepted_license":"1","department":[{"_id":"GradSch"},{"_id":"ChLa"}],"article_processing_charge":"No","related_material":{"record":[{"status":"public","id":"10802","relation":"part_of_dissertation"},{"status":"public","id":"10803","relation":"part_of_dissertation"},{"status":"public","id":"6590","relation":"part_of_dissertation"},{"id":"8724","status":"public","relation":"part_of_dissertation"}]},"language":[{"iso":"eng"}],"date_updated":"2026-04-07T14:19:48Z","ddc":["000"],"file":[{"content_type":"application/pdf","file_size":4204905,"date_updated":"2022-03-06T11:42:54Z","success":1,"relation":"main_file","file_id":"10823","checksum":"626bc523ae8822d20e635d0e2d95182e","file_name":"thesis.pdf","date_created":"2022-03-06T11:42:54Z","access_level":"open_access","creator":"nkonstan"},{"content_type":"application/x-zip-compressed","date_updated":"2022-03-10T12:11:48Z","file_size":22841103,"relation":"source_file","file_id":"10824","file_name":"thesis.zip","date_created":"2022-03-06T11:42:57Z","checksum":"e2ca2b88350ac8ea1515b948885cbcb1","creator":"nkonstan","access_level":"closed"}],"corr_author":"1","month":"03","publisher":"Institute of Science and Technology Austria","day":"08","degree_awarded":"PhD","keyword":["robustness","fairness","machine learning","PAC learning","adversarial learning"],"ec_funded":1,"year":"2022","project":[{"grant_number":"665385","name":"International IST Doctoral Program","call_identifier":"H2020","_id":"2564DBCA-B435-11E9-9278-68D0E5697425"}],"type":"dissertation"},{"ddc":["004"],"file":[{"success":1,"content_type":"application/pdf","file_size":551862,"date_updated":"2022-07-12T15:08:28Z","creator":"kschuh","access_level":"open_access","file_id":"11570","date_created":"2022-07-12T15:08:28Z","checksum":"9cac897b54a0ddf3a553a2c33e88cfda","file_name":"2022_JournalMachineLearningResearch_Konstantinov.pdf","relation":"main_file"}],"date_updated":"2026-04-07T14:19:48Z","related_material":{"record":[{"id":"13241","status":"public","relation":"shorter_version"},{"status":"public","id":"10799","relation":"dissertation_contains"}]},"language":[{"iso":"eng"}],"publication":"Journal of Machine Learning Research","has_accepted_license":"1","department":[{"_id":"ChLa"}],"article_processing_charge":"No","status":"public","external_id":{"arxiv":["2102.06004"]},"year":"2022","type":"journal_article","keyword":["Fairness","robustness","data poisoning","trustworthy machine learning","PAC learning"],"tmp":{"name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","short":"CC BY (4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","image":"/images/cc_by.png"},"day":"01","corr_author":"1","month":"05","publisher":"ML Research Press","title":"Fairness-aware PAC learning from corrupted data","oa_version":"Published Version","publication_status":"published","abstract":[{"lang":"eng","text":"Addressing fairness concerns about machine learning models is a crucial step towards their long-term adoption in real-world automated systems. While many approaches have been developed for training fair models from data, little is known about the robustness of these methods to data corruption. In this work we consider fairness-aware learning under worst-case data manipulations. We show that an adversary can in some situations force any learner to return an overly biased classifier, regardless of the sample size and with or without degrading\r\naccuracy, and that the strength of the excess bias increases for learning problems with underrepresented protected groups in the data. We also prove that our hardness results are tight up to constant factors. To this end, we study two natural learning algorithms that optimize for both accuracy and fairness and show that these algorithms enjoy guarantees that are order-optimal in terms of the corruption ratio and the protected groups frequencies in the large data\r\nlimit."}],"file_date_updated":"2022-07-12T15:08:28Z","quality_controlled":"1","page":"1-60","volume":23,"publication_identifier":{"eissn":["1533-7928"],"issn":["1532-4435"]},"article_type":"original","intvolume":"        23","acknowledgement":"The authors thank Eugenia Iofinova and Bernd Prach for providing feedback on early versions of this paper. This publication was made possible by an ETH AI Center postdoctoral fellowship to Nikola Konstantinov.","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","date_published":"2022-05-01T00:00:00Z","oa":1,"citation":{"ista":"Konstantinov NH, Lampert C. 2022. Fairness-aware PAC learning from corrupted data. Journal of Machine Learning Research. 23, 1–60.","ieee":"N. H. Konstantinov and C. Lampert, “Fairness-aware PAC learning from corrupted data,” <i>Journal of Machine Learning Research</i>, vol. 23. ML Research Press, pp. 1–60, 2022.","ama":"Konstantinov NH, Lampert C. Fairness-aware PAC learning from corrupted data. <i>Journal of Machine Learning Research</i>. 2022;23:1-60.","chicago":"Konstantinov, Nikola H, and Christoph Lampert. “Fairness-Aware PAC Learning from Corrupted Data.” <i>Journal of Machine Learning Research</i>. ML Research Press, 2022.","short":"N.H. Konstantinov, C. Lampert, Journal of Machine Learning Research 23 (2022) 1–60.","mla":"Konstantinov, Nikola H., and Christoph Lampert. “Fairness-Aware PAC Learning from Corrupted Data.” <i>Journal of Machine Learning Research</i>, vol. 23, ML Research Press, 2022, pp. 1–60.","apa":"Konstantinov, N. H., &#38; Lampert, C. (2022). Fairness-aware PAC learning from corrupted data. <i>Journal of Machine Learning Research</i>. ML Research Press."},"arxiv":1,"_id":"10802","scopus_import":"1","date_created":"2022-02-28T14:05:42Z","author":[{"id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87","full_name":"Konstantinov, Nikola H","last_name":"Konstantinov","orcid":"0009-0009-5204-7621","first_name":"Nikola H"},{"orcid":"0000-0002-4561-241X","last_name":"Lampert","first_name":"Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}]},{"article_processing_charge":"No","publication":"Computer Vision","department":[{"_id":"ChLa"}],"quality_controlled":"1","page":"1395-1397","status":"public","place":"Cham","publication_status":"published","date_updated":"2024-10-09T21:08:12Z","title":"Zero-Shot Learning","oa_version":"None","editor":[{"first_name":"Katsushi","last_name":"Ikeuchi","full_name":"Ikeuchi, Katsushi"}],"language":[{"iso":"eng"}],"abstract":[{"lang":"eng","text":"The goal of zero-shot learning is to construct a classifier that can identify object classes for which no training examples are available. When training data for some of the object classes is available but not for others, the name generalized zero-shot learning is commonly used.\r\nIn a wider sense, the phrase zero-shot is also used to describe other machine learning-based approaches that require no training data from the problem of interest, such as zero-shot action recognition or zero-shot machine translation."}],"date_published":"2021-10-13T00:00:00Z","citation":{"apa":"Lampert, C. (2021). Zero-Shot Learning. In K. Ikeuchi (Ed.), <i>Computer Vision</i> (2nd ed., pp. 1395–1397). Cham: Springer. <a href=\"https://doi.org/10.1007/978-3-030-63416-2_874\">https://doi.org/10.1007/978-3-030-63416-2_874</a>","ama":"Lampert C. Zero-Shot Learning. In: Ikeuchi K, ed. <i>Computer Vision</i>. 2nd ed. Cham: Springer; 2021:1395-1397. doi:<a href=\"https://doi.org/10.1007/978-3-030-63416-2_874\">10.1007/978-3-030-63416-2_874</a>","chicago":"Lampert, Christoph. “Zero-Shot Learning.” In <i>Computer Vision</i>, edited by Katsushi Ikeuchi, 2nd ed., 1395–97. Cham: Springer, 2021. <a href=\"https://doi.org/10.1007/978-3-030-63416-2_874\">https://doi.org/10.1007/978-3-030-63416-2_874</a>.","mla":"Lampert, Christoph. “Zero-Shot Learning.” <i>Computer Vision</i>, edited by Katsushi Ikeuchi, 2nd ed., Springer, 2021, pp. 1395–97, doi:<a href=\"https://doi.org/10.1007/978-3-030-63416-2_874\">10.1007/978-3-030-63416-2_874</a>.","short":"C. Lampert, in:, K. Ikeuchi (Ed.), Computer Vision, 2nd ed., Springer, Cham, 2021, pp. 1395–1397.","ieee":"C. Lampert, “Zero-Shot Learning,” in <i>Computer Vision</i>, 2nd ed., K. Ikeuchi, Ed. Cham: Springer, 2021, pp. 1395–1397.","ista":"Lampert C. 2021.Zero-Shot Learning. In: Computer Vision. , 1395–1397."},"day":"13","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","doi":"10.1007/978-3-030-63416-2_874","date_created":"2024-02-14T14:05:32Z","author":[{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","first_name":"Christoph","orcid":"0000-0001-8622-7887","last_name":"Lampert"}],"publisher":"Springer","corr_author":"1","_id":"14987","month":"10","type":"book_chapter","publication_identifier":{"eisbn":["9783030634162"],"isbn":["9783030634155"]},"year":"2021","edition":"2"},{"user_id":"317138e5-6ab7-11ef-aa6d-ffef3953e345","date_published":"2021-03-17T00:00:00Z","oa":1,"citation":{"ieee":"V. Volhejn and C. Lampert, “Does SGD implicitly optimize for smoothness?,” in <i>42nd German Conference on Pattern Recognition</i>, Tübingen, Germany, 2021, vol. 12544, pp. 246–259.","ista":"Volhejn V, Lampert C. 2021. Does SGD implicitly optimize for smoothness? 42nd German Conference on Pattern Recognition. DAGM GCPR: German Conference on Pattern Recognition LNCS vol. 12544, 246–259.","apa":"Volhejn, V., &#38; Lampert, C. (2021). Does SGD implicitly optimize for smoothness? In <i>42nd German Conference on Pattern Recognition</i> (Vol. 12544, pp. 246–259). Tübingen, Germany: Springer. <a href=\"https://doi.org/10.1007/978-3-030-71278-5_18\">https://doi.org/10.1007/978-3-030-71278-5_18</a>","ama":"Volhejn V, Lampert C. Does SGD implicitly optimize for smoothness? In: <i>42nd German Conference on Pattern Recognition</i>. Vol 12544. LNCS. Springer; 2021:246-259. doi:<a href=\"https://doi.org/10.1007/978-3-030-71278-5_18\">10.1007/978-3-030-71278-5_18</a>","chicago":"Volhejn, Vaclav, and Christoph Lampert. “Does SGD Implicitly Optimize for Smoothness?” In <i>42nd German Conference on Pattern Recognition</i>, 12544:246–59. LNCS. Springer, 2021. <a href=\"https://doi.org/10.1007/978-3-030-71278-5_18\">https://doi.org/10.1007/978-3-030-71278-5_18</a>.","mla":"Volhejn, Vaclav, and Christoph Lampert. “Does SGD Implicitly Optimize for Smoothness?” <i>42nd German Conference on Pattern Recognition</i>, vol. 12544, Springer, 2021, pp. 246–59, doi:<a href=\"https://doi.org/10.1007/978-3-030-71278-5_18\">10.1007/978-3-030-71278-5_18</a>.","short":"V. Volhejn, C. Lampert, in:, 42nd German Conference on Pattern Recognition, Springer, 2021, pp. 246–259."},"_id":"9210","scopus_import":"1","author":[{"last_name":"Volhejn","first_name":"Vaclav","id":"d5235fb4-7a6d-11eb-b254-f25d12d631a8","full_name":"Volhejn, Vaclav"},{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","first_name":"Christoph","orcid":"0000-0001-8622-7887","last_name":"Lampert"}],"date_created":"2021-03-01T09:01:16Z","doi":"10.1007/978-3-030-71278-5_18","publication_identifier":{"isbn":["9783030712778"],"eissn":["1611-3349"],"issn":["0302-9743"]},"volume":12544,"intvolume":"     12544","quality_controlled":"1","file_date_updated":"2022-08-12T07:27:58Z","page":"246-259","title":"Does SGD implicitly optimize for smoothness?","oa_version":"Submitted Version","publication_status":"published","abstract":[{"text":"Modern neural networks can easily fit their training set perfectly. Surprisingly, despite being “overfit” in this way, they tend to generalize well to future data, thereby defying the classic bias–variance trade-off of machine learning theory. Of the many possible explanations, a prevalent one is that training by stochastic gradient descent (SGD) imposes an implicit bias that leads it to learn simple functions, and these simple functions generalize well. However, the specifics of this implicit bias are not well understood.\r\nIn this work, we explore the smoothness conjecture which states that SGD is implicitly biased towards learning functions that are smooth. We propose several measures to formalize the intuitive notion of smoothness, and we conduct experiments to determine whether SGD indeed implicitly optimizes for these measures. Our findings rule out the possibility that smoothness measures based on first-order derivatives are being implicitly enforced. They are supportive, though, of the smoothness conjecture for measures based on second-order derivatives.","lang":"eng"}],"day":"17","month":"03","conference":{"location":"Tübingen, Germany","end_date":"2020-10-01","start_date":"2020-09-28","name":"DAGM GCPR: German Conference on Pattern Recognition "},"publisher":"Springer","year":"2021","type":"conference","publication":"42nd German Conference on Pattern Recognition","isi":1,"has_accepted_license":"1","department":[{"_id":"ChLa"}],"article_processing_charge":"No","status":"public","external_id":{"isi":["001500603200018"]},"series_title":"LNCS","date_updated":"2025-09-10T10:00:33Z","ddc":["510"],"file":[{"success":1,"content_type":"application/pdf","date_updated":"2022-08-12T07:27:58Z","file_size":420234,"relation":"main_file","creator":"dernst","access_level":"open_access","date_created":"2022-08-12T07:27:58Z","file_name":"2020_GCPR_submitted_Volhejn.pdf","checksum":"3e3628ab1cf658d82524963f808004ea","file_id":"11820"}],"language":[{"iso":"eng"}]},{"abstract":[{"lang":"eng","text":"Given the abundance of applications of ranking in recent years, addressing fairness concerns around automated ranking systems becomes necessary for increasing the trust among end-users. Previous work on fair ranking has mostly focused on application-specific fairness notions, often tailored to online advertising, and it rarely considers learning as part of the process. In this work, we show how to transfer numerous fairness notions from binary classification to a learning to rank setting. Our formalism allows us to design methods for incorporating fairness objectives with provable generalization guarantees. An extensive experimental evaluation shows that our method can improve ranking fairness substantially with no or only little loss of model quality."}],"related_material":{"record":[{"status":"public","id":"10799","relation":"dissertation_contains"}]},"language":[{"iso":"eng"}],"oa_version":"Preprint","title":"Fairness through regularization for learning to rank","date_updated":"2026-04-07T14:19:48Z","publication_status":"draft","main_file_link":[{"url":"https://arxiv.org/abs/2102.05996","open_access":"1"}],"status":"public","external_id":{"arxiv":["2102.05996"]},"article_number":"2102.05996","publication":"arXiv","department":[{"_id":"ChLa"}],"article_processing_charge":"No","year":"2021","type":"preprint","corr_author":"1","_id":"10803","month":"06","doi":"10.48550/arXiv.2102.05996","author":[{"id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87","full_name":"Konstantinov, Nikola H","last_name":"Konstantinov","orcid":"0009-0009-5204-7621","first_name":"Nikola H"},{"orcid":"0000-0002-4561-241X","last_name":"Lampert","first_name":"Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"date_created":"2022-02-28T14:13:59Z","day":"07","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","date_published":"2021-06-07T00:00:00Z","oa":1,"citation":{"ieee":"N. H. Konstantinov and C. Lampert, “Fairness through regularization for learning to rank,” <i>arXiv</i>. .","ista":"Konstantinov NH, Lampert C. Fairness through regularization for learning to rank. arXiv, 2102.05996.","apa":"Konstantinov, N. H., &#38; Lampert, C. (n.d.). Fairness through regularization for learning to rank. <i>arXiv</i>. <a href=\"https://doi.org/10.48550/arXiv.2102.05996\">https://doi.org/10.48550/arXiv.2102.05996</a>","ama":"Konstantinov NH, Lampert C. Fairness through regularization for learning to rank. <i>arXiv</i>. doi:<a href=\"https://doi.org/10.48550/arXiv.2102.05996\">10.48550/arXiv.2102.05996</a>","short":"N.H. Konstantinov, C. Lampert, ArXiv (n.d.).","mla":"Konstantinov, Nikola H., and Christoph Lampert. “Fairness through Regularization for Learning to Rank.” <i>ArXiv</i>, 2102.05996, doi:<a href=\"https://doi.org/10.48550/arXiv.2102.05996\">10.48550/arXiv.2102.05996</a>.","chicago":"Konstantinov, Nikola H, and Christoph Lampert. “Fairness through Regularization for Learning to Rank.” <i>ArXiv</i>, n.d. <a href=\"https://doi.org/10.48550/arXiv.2102.05996\">https://doi.org/10.48550/arXiv.2102.05996</a>."},"arxiv":1},{"department":[{"_id":"GradSch"},{"_id":"ChLa"}],"has_accepted_license":"1","article_processing_charge":"No","status":"public","ddc":["000"],"file":[{"success":1,"content_type":"application/pdf","file_size":2673905,"date_updated":"2021-05-24T11:22:29Z","access_level":"open_access","creator":"bphuong","checksum":"4f0abe64114cfed264f9d36e8d1197e3","file_name":"mph-thesis-v519-pdfimages.pdf","file_id":"9419","date_created":"2021-05-24T11:22:29Z","relation":"main_file"},{"relation":"source_file","access_level":"closed","creator":"bphuong","date_created":"2021-05-24T11:56:02Z","checksum":"f5699e876bc770a9b0df8345a77720a2","file_id":"9420","file_name":"thesis.zip","content_type":"application/zip","date_updated":"2021-05-24T11:56:02Z","file_size":92995100}],"date_updated":"2026-04-08T07:01:17Z","language":[{"iso":"eng"}],"related_material":{"record":[{"relation":"part_of_dissertation","id":"7435","status":"deleted"},{"id":"7481","status":"public","relation":"part_of_dissertation"},{"id":"9416","status":"public","relation":"part_of_dissertation"},{"id":"7479","status":"public","relation":"part_of_dissertation"}]},"day":"30","degree_awarded":"PhD","month":"05","corr_author":"1","publisher":"Institute of Science and Technology Austria","year":"2021","type":"dissertation","file_date_updated":"2021-05-24T11:56:02Z","page":"125","title":"Underspecification in deep learning","oa_version":"Published Version","publication_status":"published","abstract":[{"lang":"eng","text":"Deep learning is best known for its empirical success across a wide range of applications\r\nspanning computer vision, natural language processing and speech. Of equal significance,\r\nthough perhaps less known, are its ramifications for learning theory: deep networks have\r\nbeen observed to perform surprisingly well in the high-capacity regime, aka the overfitting\r\nor underspecified regime. Classically, this regime on the far right of the bias-variance curve\r\nis associated with poor generalisation; however, recent experiments with deep networks\r\nchallenge this view.\r\n\r\nThis thesis is devoted to investigating various aspects of underspecification in deep learning.\r\nFirst, we argue that deep learning models are underspecified on two levels: a) any given\r\ntraining dataset can be fit by many different functions, and b) any given function can be\r\nexpressed by many different parameter configurations. We refer to the second kind of\r\nunderspecification as parameterisation redundancy and we precisely characterise its extent.\r\nSecond, we characterise the implicit criteria (the inductive bias) that guide learning in the\r\nunderspecified regime. Specifically, we consider a nonlinear but tractable classification\r\nsetting, and show that given the choice, neural networks learn classifiers with a large margin.\r\nThird, we consider learning scenarios where the inductive bias is not by itself sufficient to\r\ndeal with underspecification. We then study different ways of ‘tightening the specification’: i)\r\nIn the setting of representation learning with variational autoencoders, we propose a hand-\r\ncrafted regulariser based on mutual information. ii) In the setting of binary classification, we\r\nconsider soft-label (real-valued) supervision. We derive a generalisation bound for linear\r\nnetworks supervised in this way and verify that soft labels facilitate fast learning. Finally, we\r\nexplore an application of soft-label supervision to the training of multi-exit models."}],"acknowledged_ssus":[{"_id":"ScienComp"},{"_id":"CampIT"},{"_id":"E-Lib"}],"user_id":"ba8df636-2132-11f1-aed0-ed93e2281fdd","citation":{"ista":"Phuong M. 2021. Underspecification in deep learning. Institute of Science and Technology Austria.","ieee":"M. Phuong, “Underspecification in deep learning,” Institute of Science and Technology Austria, 2021.","ama":"Phuong M. Underspecification in deep learning. 2021. doi:<a href=\"https://doi.org/10.15479/AT:ISTA:9418\">10.15479/AT:ISTA:9418</a>","short":"M. Phuong, Underspecification in Deep Learning, Institute of Science and Technology Austria, 2021.","chicago":"Phuong, Mary. “Underspecification in Deep Learning.” Institute of Science and Technology Austria, 2021. <a href=\"https://doi.org/10.15479/AT:ISTA:9418\">https://doi.org/10.15479/AT:ISTA:9418</a>.","mla":"Phuong, Mary. <i>Underspecification in Deep Learning</i>. Institute of Science and Technology Austria, 2021, doi:<a href=\"https://doi.org/10.15479/AT:ISTA:9418\">10.15479/AT:ISTA:9418</a>.","apa":"Phuong, M. (2021). <i>Underspecification in deep learning</i>. Institute of Science and Technology Austria. <a href=\"https://doi.org/10.15479/AT:ISTA:9418\">https://doi.org/10.15479/AT:ISTA:9418</a>"},"supervisor":[{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph","last_name":"Lampert","orcid":"0000-0001-8622-7887","first_name":"Christoph"}],"oa":1,"date_published":"2021-05-30T00:00:00Z","OA_place":"publisher","_id":"9418","doi":"10.15479/AT:ISTA:9418","date_created":"2021-05-24T13:06:23Z","author":[{"id":"3EC6EE64-F248-11E8-B48F-1D18A9856A87","full_name":"Bui Thi Mai, Phuong","last_name":"Bui Thi Mai","first_name":"Phuong"}],"alternative_title":["ISTA Thesis"],"publication_identifier":{"issn":["2663-337X"]}},{"language":[{"iso":"eng"}],"related_material":{"record":[{"id":"9418","status":"public","relation":"dissertation_contains"}]},"file":[{"content_type":"application/pdf","date_updated":"2021-05-24T11:15:57Z","file_size":502356,"relation":"main_file","creator":"bphuong","access_level":"open_access","checksum":"f34ff17017527db5ba6927f817bdd125","date_created":"2021-05-24T11:15:57Z","file_name":"iclr2021_conference.pdf","file_id":"9417"}],"date_updated":"2026-04-08T07:01:16Z","ddc":["000"],"status":"public","article_processing_charge":"No","department":[{"_id":"GradSch"},{"_id":"ChLa"}],"has_accepted_license":"1","publication":"9th International Conference on Learning Representations","type":"conference","year":"2021","conference":{"location":"Virtual","end_date":"2021-05-07","start_date":"2021-05-03","name":"ICLR: International Conference on Learning Representations"},"month":"05","corr_author":"1","day":"01","abstract":[{"lang":"eng","text":"We study the inductive bias of two-layer ReLU networks trained by gradient flow. We identify a class of easy-to-learn (`orthogonally separable') datasets, and characterise the solution that ReLU networks trained on such datasets converge to. Irrespective of network width, the solution turns out to be a combination of two max-margin classifiers: one corresponding to the positive data subset and one corresponding to the negative data subset. The proof is based on the recently introduced concept of extremal sectors, for which we prove a number of properties in the context of orthogonal separability. In particular, we prove stationarity of activation patterns from some time  onwards, which enables a reduction of the ReLU network to an ensemble of linear subnetworks."}],"publication_status":"published","title":"The inductive bias of ReLU networks on orthogonally separable data","oa_version":"Published Version","main_file_link":[{"open_access":"1","url":"https://openreview.net/pdf?id=krz7T0xU9Z_"}],"quality_controlled":"1","file_date_updated":"2021-05-24T11:15:57Z","author":[{"full_name":"Bui Thi Mai, Phuong","id":"3EC6EE64-F248-11E8-B48F-1D18A9856A87","first_name":"Phuong","last_name":"Bui Thi Mai"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887","last_name":"Lampert","first_name":"Christoph"}],"date_created":"2021-05-24T11:16:46Z","scopus_import":"1","_id":"9416","citation":{"ista":"Phuong M, Lampert C. 2021. The inductive bias of ReLU networks on orthogonally separable data. 9th International Conference on Learning Representations. ICLR: International Conference on Learning Representations.","ieee":"M. Phuong and C. Lampert, “The inductive bias of ReLU networks on orthogonally separable data,” in <i>9th International Conference on Learning Representations</i>, Virtual, 2021.","ama":"Phuong M, Lampert C. The inductive bias of ReLU networks on orthogonally separable data. In: <i>9th International Conference on Learning Representations</i>. ; 2021.","short":"M. Phuong, C. Lampert, in:, 9th International Conference on Learning Representations, 2021.","chicago":"Phuong, Mary, and Christoph Lampert. “The Inductive Bias of ReLU Networks on Orthogonally Separable Data.” In <i>9th International Conference on Learning Representations</i>, 2021.","mla":"Phuong, Mary, and Christoph Lampert. “The Inductive Bias of ReLU Networks on Orthogonally Separable Data.” <i>9th International Conference on Learning Representations</i>, 2021.","apa":"Phuong, M., &#38; Lampert, C. (2021). The inductive bias of ReLU networks on orthogonally separable data. In <i>9th International Conference on Learning Representations</i>. Virtual."},"date_published":"2021-05-01T00:00:00Z","oa":1,"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87"},{"author":[{"first_name":"Titas","last_name":"Anciukevicius","full_name":"Anciukevicius, Titas"},{"first_name":"Christoph","orcid":"0000-0001-8622-7887","last_name":"Lampert","full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"},{"first_name":"Paul M","last_name":"Henderson","orcid":"0000-0002-5198-7445","full_name":"Henderson, Paul M","id":"13C09E74-18D9-11E9-8878-32CFE5697425"}],"date_created":"2020-06-29T23:55:23Z","doi":"10.48550/arXiv.2004.00642","month":"04","_id":"8063","arxiv":1,"citation":{"apa":"Anciukevicius, T., Lampert, C., &#38; Henderson, P. M. (n.d.). Object-centric image generation with factored depths, locations, and appearances. <i>arXiv</i>. <a href=\"https://doi.org/10.48550/arXiv.2004.00642\">https://doi.org/10.48550/arXiv.2004.00642</a>","ama":"Anciukevicius T, Lampert C, Henderson PM. Object-centric image generation with factored depths, locations, and appearances. <i>arXiv</i>. doi:<a href=\"https://doi.org/10.48550/arXiv.2004.00642\">10.48550/arXiv.2004.00642</a>","chicago":"Anciukevicius, Titas, Christoph Lampert, and Paul M Henderson. “Object-Centric Image Generation with Factored Depths, Locations, and Appearances.” <i>ArXiv</i>, n.d. <a href=\"https://doi.org/10.48550/arXiv.2004.00642\">https://doi.org/10.48550/arXiv.2004.00642</a>.","mla":"Anciukevicius, Titas, et al. “Object-Centric Image Generation with Factored Depths, Locations, and Appearances.” <i>ArXiv</i>, 2004.00642, doi:<a href=\"https://doi.org/10.48550/arXiv.2004.00642\">10.48550/arXiv.2004.00642</a>.","short":"T. Anciukevicius, C. Lampert, P.M. Henderson, ArXiv (n.d.).","ieee":"T. Anciukevicius, C. Lampert, and P. M. Henderson, “Object-centric image generation with factored depths, locations, and appearances,” <i>arXiv</i>. .","ista":"Anciukevicius T, Lampert C, Henderson PM. Object-centric image generation with factored depths, locations, and appearances. arXiv, 2004.00642."},"oa":1,"date_published":"2020-04-01T00:00:00Z","license":"https://creativecommons.org/licenses/by-sa/4.0/","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","day":"01","tmp":{"short":"CC BY-SA (4.0)","name":"Creative Commons Attribution-ShareAlike 4.0 International Public License (CC BY-SA 4.0)","legal_code_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","image":"/images/cc_by_sa.png"},"type":"preprint","year":"2020","article_number":"2004.00642","external_id":{"arxiv":["2004.00642"]},"status":"public","main_file_link":[{"url":"https://arxiv.org/abs/2004.00642","open_access":"1"}],"article_processing_charge":"No","department":[{"_id":"ChLa"}],"publication":"arXiv","language":[{"iso":"eng"}],"abstract":[{"lang":"eng","text":"We present a generative model of images that explicitly reasons over the set\r\nof objects they show. Our model learns a structured latent representation that\r\nseparates objects from each other and from the background; unlike prior works,\r\nit explicitly represents the 2D position and depth of each object, as well as\r\nan embedding of its segmentation mask and appearance. The model can be trained\r\nfrom images alone in a purely unsupervised fashion without the need for object\r\nmasks or depth information. Moreover, it always generates complete objects,\r\neven though a significant fraction of training images contain occlusions.\r\nFinally, we show that our model can infer decompositions of novel images into\r\ntheir constituent objects, including accurate prediction of depth ordering and\r\nsegmentation of occluded parts."}],"date_updated":"2025-01-20T14:20:49Z","publication_status":"submitted","ddc":["004"],"title":"Object-centric image generation with factored depths, locations, and appearances","oa_version":"Preprint"},{"ddc":["004"],"date_updated":"2023-10-17T07:37:11Z","file":[{"file_name":"paper.pdf","date_created":"2020-07-31T16:57:12Z","file_id":"8187","access_level":"open_access","creator":"phenders","relation":"main_file","file_size":10262773,"date_updated":"2020-07-31T16:57:12Z","content_type":"application/pdf","success":1}],"language":[{"iso":"eng"}],"article_processing_charge":"No","department":[{"_id":"ChLa"}],"has_accepted_license":"1","publication":"Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition","external_id":{"arxiv":["2004.04180"]},"status":"public","type":"conference","year":"2020","day":"01","publisher":"IEEE","conference":{"end_date":"2020-06-19","name":"CVPR: Conference on Computer Vision and Pattern Recognition","start_date":"2020-06-14","location":"Virtual"},"month":"07","publication_status":"published","title":"Leveraging 2D data to learn textured 3D mesh generation","oa_version":"Submitted Version","abstract":[{"lang":"eng","text":"Numerous methods have been proposed for probabilistic generative modelling of\r\n3D objects. However, none of these is able to produce textured objects, which\r\nrenders them of limited use for practical tasks. In this work, we present the\r\nfirst generative model of textured 3D meshes. Training such a model would\r\ntraditionally require a large dataset of textured meshes, but unfortunately,\r\nexisting datasets of meshes lack detailed textures. We instead propose a new\r\ntraining methodology that allows learning from collections of 2D images without\r\nany 3D information. To do so, we train our model to explain a distribution of\r\nimages by modelling each image as a 3D foreground object placed in front of a\r\n2D background. Thus, it learns to generate meshes that when rendered, produce\r\nimages similar to those in its training set.\r\n  A well-known problem when generating meshes with deep networks is the\r\nemergence of self-intersections, which are problematic for many use-cases. As a\r\nsecond contribution we therefore introduce a new generation process for 3D\r\nmeshes that guarantees no self-intersections arise, based on the physical\r\nintuition that faces should push one another out of the way as they move.\r\n  We conduct extensive experiments on our approach, reporting quantitative and\r\nqualitative results on both synthetic data and natural images. These show our\r\nmethod successfully learns to generate plausible and diverse textured 3D\r\nsamples for five challenging object classes."}],"quality_controlled":"1","file_date_updated":"2020-07-31T16:57:12Z","page":"7498-7507","main_file_link":[{"url":"https://openaccess.thecvf.com/content_CVPR_2020/papers/Henderson_Leveraging_2D_Data_to_Learn_Textured_3D_Mesh_Generation_CVPR_2020_paper.pdf","open_access":"1"}],"publication_identifier":{"eissn":["2575-7075"],"eisbn":["9781728171685"]},"arxiv":1,"citation":{"apa":"Henderson, P. M., Tsiminaki, V., &#38; Lampert, C. (2020). Leveraging 2D data to learn textured 3D mesh generation. In <i>Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition</i> (pp. 7498–7507). Virtual: IEEE. <a href=\"https://doi.org/10.1109/CVPR42600.2020.00752\">https://doi.org/10.1109/CVPR42600.2020.00752</a>","mla":"Henderson, Paul M., et al. “Leveraging 2D Data to Learn Textured 3D Mesh Generation.” <i>Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>, IEEE, 2020, pp. 7498–507, doi:<a href=\"https://doi.org/10.1109/CVPR42600.2020.00752\">10.1109/CVPR42600.2020.00752</a>.","short":"P.M. Henderson, V. Tsiminaki, C. Lampert, in:, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, 2020, pp. 7498–7507.","chicago":"Henderson, Paul M, Vagia Tsiminaki, and Christoph Lampert. “Leveraging 2D Data to Learn Textured 3D Mesh Generation.” In <i>Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>, 7498–7507. IEEE, 2020. <a href=\"https://doi.org/10.1109/CVPR42600.2020.00752\">https://doi.org/10.1109/CVPR42600.2020.00752</a>.","ama":"Henderson PM, Tsiminaki V, Lampert C. Leveraging 2D data to learn textured 3D mesh generation. In: <i>Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>. IEEE; 2020:7498-7507. doi:<a href=\"https://doi.org/10.1109/CVPR42600.2020.00752\">10.1109/CVPR42600.2020.00752</a>","ieee":"P. M. Henderson, V. Tsiminaki, and C. Lampert, “Leveraging 2D data to learn textured 3D mesh generation,” in <i>Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>, Virtual, 2020, pp. 7498–7507.","ista":"Henderson PM, Tsiminaki V, Lampert C. 2020. Leveraging 2D data to learn textured 3D mesh generation. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. CVPR: Conference on Computer Vision and Pattern Recognition, 7498–7507."},"date_published":"2020-07-01T00:00:00Z","oa":1,"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","date_created":"2020-07-31T16:53:49Z","doi":"10.1109/CVPR42600.2020.00752","author":[{"first_name":"Paul M","orcid":"0000-0002-5198-7445","last_name":"Henderson","full_name":"Henderson, Paul M","id":"13C09E74-18D9-11E9-8878-32CFE5697425"},{"last_name":"Tsiminaki","first_name":"Vagia","full_name":"Tsiminaki, Vagia"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph","last_name":"Lampert","orcid":"0000-0001-8622-7887","first_name":"Christoph"}],"scopus_import":"1","_id":"8186"},{"acknowledged_ssus":[{"_id":"ScienComp"}],"abstract":[{"text":"A natural approach to generative modeling of videos is to represent them as a composition of moving objects. Recent works model a set of 2D sprites over a slowly-varying background, but without considering the underlying 3D scene that\r\ngives rise to them. We instead propose to model a video as the view seen while moving through a scene with multiple 3D objects and a 3D background. Our model is trained from monocular videos without any supervision, yet learns to\r\ngenerate coherent 3D scenes containing several moving objects. We conduct detailed experiments on two datasets, going beyond the visual complexity supported by state-of-the-art generative approaches. We evaluate our method on\r\ndepth-prediction and 3D object detection---tasks which cannot be addressed by those earlier works---and show it out-performs them even on 2D instance segmentation and tracking.","lang":"eng"}],"publication_status":"published","oa_version":"Preprint","title":"Unsupervised object-centric video generation and decomposition in 3D","page":"3106–3117","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2007.06705"}],"quality_controlled":"1","intvolume":"        33","acknowledgement":"This research was supported by the Scientific Service Units (SSU) of IST Austria through resources\r\nprovided by Scientific Computing (SciComp). PH is employed part-time by Blackford Analysis, but\r\nthey did not support this project in any way.","volume":33,"publication_identifier":{"isbn":["9781713829546"]},"alternative_title":["Advances in Neural Information Processing Systems"],"date_created":"2020-07-31T16:59:19Z","author":[{"first_name":"Paul M","orcid":"0000-0002-5198-7445","last_name":"Henderson","full_name":"Henderson, Paul M","id":"13C09E74-18D9-11E9-8878-32CFE5697425"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887","last_name":"Lampert","first_name":"Christoph"}],"_id":"8188","date_published":"2020-07-07T00:00:00Z","oa":1,"arxiv":1,"citation":{"ista":"Henderson PM, Lampert C. 2020. Unsupervised object-centric video generation and decomposition in 3D. 34th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 33, 3106–3117.","ieee":"P. M. Henderson and C. Lampert, “Unsupervised object-centric video generation and decomposition in 3D,” in <i>34th Conference on Neural Information Processing Systems</i>, Vancouver, Canada, 2020, vol. 33, pp. 3106–3117.","short":"P.M. Henderson, C. Lampert, in:, 34th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2020, pp. 3106–3117.","chicago":"Henderson, Paul M, and Christoph Lampert. “Unsupervised Object-Centric Video Generation and Decomposition in 3D.” In <i>34th Conference on Neural Information Processing Systems</i>, 33:3106–3117. Neural Information Processing Systems Foundation, 2020.","mla":"Henderson, Paul M., and Christoph Lampert. “Unsupervised Object-Centric Video Generation and Decomposition in 3D.” <i>34th Conference on Neural Information Processing Systems</i>, vol. 33, Neural Information Processing Systems Foundation, 2020, pp. 3106–3117.","ama":"Henderson PM, Lampert C. Unsupervised object-centric video generation and decomposition in 3D. In: <i>34th Conference on Neural Information Processing Systems</i>. Vol 33. Neural Information Processing Systems Foundation; 2020:3106–3117.","apa":"Henderson, P. M., &#38; Lampert, C. (2020). Unsupervised object-centric video generation and decomposition in 3D. In <i>34th Conference on Neural Information Processing Systems</i> (Vol. 33, pp. 3106–3117). Vancouver, Canada: Neural Information Processing Systems Foundation."},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","language":[{"iso":"eng"}],"date_updated":"2025-05-14T11:26:57Z","external_id":{"arxiv":["2007.06705"]},"status":"public","article_processing_charge":"No","publication":"34th Conference on Neural Information Processing Systems","department":[{"_id":"ChLa"}],"type":"conference","year":"2020","conference":{"location":"Vancouver, Canada","start_date":"2020-12-06","name":"NeurIPS: Neural Information Processing Systems","end_date":"2020-12-12"},"publisher":"Neural Information Processing Systems Foundation","corr_author":"1","month":"07","day":"07"},{"ddc":["004"],"date_updated":"2025-04-15T07:10:25Z","file":[{"file_size":1715072,"date_updated":"2020-07-14T12:47:45Z","content_type":"application/pdf","relation":"main_file","file_name":"2019_IJCV_Sun.pdf","date_created":"2019-11-26T10:30:02Z","file_id":"7110","checksum":"155e63edf664dcacb3bdc1c2223e606f","access_level":"open_access","creator":"dernst"}],"related_material":{"record":[{"status":"public","id":"6482","relation":"earlier_version"}],"link":[{"relation":"erratum","url":"https://doi.org/10.1007/s11263-019-01262-5"}]},"language":[{"iso":"eng"}],"publication":"International Journal of Computer Vision","isi":1,"department":[{"_id":"ChLa"}],"has_accepted_license":"1","article_processing_charge":"Yes (via OA deal)","status":"public","external_id":{"isi":["000494406800001"]},"year":"2020","project":[{"call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding","grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425"},{"_id":"B67AFEDC-15C9-11EA-A837-991A96BB2854","name":"IST Austria Open Access Fund"}],"type":"journal_article","tmp":{"name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","short":"CC BY (4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","image":"/images/cc_by.png"},"ec_funded":1,"day":"01","issue":"4","corr_author":"1","month":"04","publisher":"Springer Nature","oa_version":"Published Version","title":"KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications","publication_status":"published","abstract":[{"lang":"eng","text":"We study the problem of automatically detecting if a given multi-class classifier operates outside of its specifications (out-of-specs), i.e. on input data from a different distribution than what it was trained for. This is an important problem to solve on the road towards creating reliable computer vision systems for real-world applications, because the quality of a classifier’s predictions cannot be guaranteed if it operates out-of-specs. Previously proposed methods for out-of-specs detection make decisions on the level of single inputs. This, however, is insufficient to achieve low false positive rate and high false negative rates at the same time. In this work, we describe a new procedure named KS(conf), based on statistical reasoning. Its main component is a classical Kolmogorov–Smirnov test that is applied to the set of predicted confidence values for batches of samples. Working with batches instead of single samples allows increasing the true positive rate without negatively affecting the false positive rate, thereby overcoming a crucial limitation of single sample tests. We show by extensive experiments using a variety of convolutional network architectures and datasets that KS(conf) reliably detects out-of-specs situations even under conditions where other tests fail. It furthermore has a number of properties that make it an excellent candidate for practical deployment: it is easy to implement, adds almost no overhead to the system, works with any classifier that outputs confidence scores, and requires no a priori knowledge about how the data distribution could change."}],"quality_controlled":"1","file_date_updated":"2020-07-14T12:47:45Z","page":"970-995","publication_identifier":{"issn":["0920-5691"],"eissn":["1573-1405"]},"volume":128,"article_type":"original","intvolume":"       128","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","date_published":"2020-04-01T00:00:00Z","oa":1,"citation":{"mla":"Sun, Rémy, and Christoph Lampert. “KS(Conf): A Light-Weight Test If a Multiclass Classifier Operates Outside of Its Specifications.” <i>International Journal of Computer Vision</i>, vol. 128, no. 4, Springer Nature, 2020, pp. 970–95, doi:<a href=\"https://doi.org/10.1007/s11263-019-01232-x\">10.1007/s11263-019-01232-x</a>.","chicago":"Sun, Rémy, and Christoph Lampert. “KS(Conf): A Light-Weight Test If a Multiclass Classifier Operates Outside of Its Specifications.” <i>International Journal of Computer Vision</i>. Springer Nature, 2020. <a href=\"https://doi.org/10.1007/s11263-019-01232-x\">https://doi.org/10.1007/s11263-019-01232-x</a>.","short":"R. Sun, C. Lampert, International Journal of Computer Vision 128 (2020) 970–995.","ama":"Sun R, Lampert C. KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications. <i>International Journal of Computer Vision</i>. 2020;128(4):970-995. doi:<a href=\"https://doi.org/10.1007/s11263-019-01232-x\">10.1007/s11263-019-01232-x</a>","apa":"Sun, R., &#38; Lampert, C. (2020). KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications. <i>International Journal of Computer Vision</i>. Springer Nature. <a href=\"https://doi.org/10.1007/s11263-019-01232-x\">https://doi.org/10.1007/s11263-019-01232-x</a>","ista":"Sun R, Lampert C. 2020. KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications. International Journal of Computer Vision. 128(4), 970–995.","ieee":"R. Sun and C. Lampert, “KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications,” <i>International Journal of Computer Vision</i>, vol. 128, no. 4. Springer Nature, pp. 970–995, 2020."},"_id":"6944","scopus_import":"1","doi":"10.1007/s11263-019-01232-x","author":[{"first_name":"Rémy","last_name":"Sun","full_name":"Sun, Rémy"},{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","first_name":"Christoph","orcid":"0000-0001-8622-7887","last_name":"Lampert"}],"date_created":"2019-10-14T09:14:28Z"},{"status":"public","external_id":{"isi":["000491042100002"],"arxiv":["1901.06447"]},"isi":1,"has_accepted_license":"1","department":[{"_id":"ChLa"}],"publication":"International Journal of Computer Vision","article_processing_charge":"Yes (via OA deal)","language":[{"iso":"eng"}],"date_updated":"2025-04-15T06:53:15Z","ddc":["004"],"file":[{"file_id":"6973","date_created":"2019-10-25T10:28:29Z","checksum":"a0f05dd4f5f64e4f713d8d9d4b5b1e3f","file_name":"2019_CompVision_Henderson.pdf","access_level":"open_access","creator":"dernst","relation":"main_file","date_updated":"2020-07-14T12:47:46Z","file_size":2243134,"content_type":"application/pdf"}],"month":"04","corr_author":"1","publisher":"Springer Nature","day":"01","tmp":{"name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","short":"CC BY (4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","image":"/images/cc_by.png"},"year":"2020","type":"journal_article","project":[{"name":"IST Austria Open Access Fund","_id":"B67AFEDC-15C9-11EA-A837-991A96BB2854"}],"page":"835-854","file_date_updated":"2020-07-14T12:47:46Z","quality_controlled":"1","abstract":[{"text":"We present a unified framework tackling two problems: class-specific 3D reconstruction from a single image, and generation of new 3D shape samples. These tasks have received considerable attention recently; however, most existing approaches rely on 3D supervision, annotation of 2D images with keypoints or poses, and/or training with multiple views of each object instance. Our framework is very general: it can be trained in similar settings to existing approaches, while also supporting weaker supervision. Importantly, it can be trained purely from 2D images, without pose annotations, and with only a single view per instance. We employ meshes as an output representation, instead of voxels used in most prior work. This allows us to reason over lighting parameters and exploit shading information during training, which previous 2D-supervised methods cannot. Thus, our method can learn to generate and reconstruct concave object classes. We evaluate our approach in various settings, showing that: (i) it learns to disentangle shape from pose and lighting; (ii) using shading in the loss improves performance compared to just silhouettes; (iii) when using a standard single white light, our model outperforms state-of-the-art 2D-supervised methods, both with and without pose supervision, thanks to exploiting shading cues; (iv) performance improves further when using multiple coloured lights, even approaching that of state-of-the-art 3D-supervised methods; (v) shapes produced by our model capture smooth surfaces and fine details better than voxel-based approaches; and (vi) our approach supports concave classes such as bathtubs and sofas, which methods based on silhouettes cannot learn.","lang":"eng"}],"oa_version":"Published Version","title":"Learning single-image 3D reconstruction by generative modelling of shape, pose and shading","publication_status":"published","scopus_import":"1","_id":"6952","doi":"10.1007/s11263-019-01219-8","date_created":"2019-10-17T13:38:20Z","author":[{"last_name":"Henderson","orcid":"0000-0002-5198-7445","first_name":"Paul M","id":"13C09E74-18D9-11E9-8878-32CFE5697425","full_name":"Henderson, Paul M"},{"last_name":"Ferrari","first_name":"Vittorio","full_name":"Ferrari, Vittorio"}],"user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","arxiv":1,"citation":{"apa":"Henderson, P. M., &#38; Ferrari, V. (2020). Learning single-image 3D reconstruction by generative modelling of shape, pose and shading. <i>International Journal of Computer Vision</i>. Springer Nature. <a href=\"https://doi.org/10.1007/s11263-019-01219-8\">https://doi.org/10.1007/s11263-019-01219-8</a>","short":"P.M. Henderson, V. Ferrari, International Journal of Computer Vision 128 (2020) 835–854.","mla":"Henderson, Paul M., and Vittorio Ferrari. “Learning Single-Image 3D Reconstruction by Generative Modelling of Shape, Pose and Shading.” <i>International Journal of Computer Vision</i>, vol. 128, Springer Nature, 2020, pp. 835–54, doi:<a href=\"https://doi.org/10.1007/s11263-019-01219-8\">10.1007/s11263-019-01219-8</a>.","chicago":"Henderson, Paul M, and Vittorio Ferrari. “Learning Single-Image 3D Reconstruction by Generative Modelling of Shape, Pose and Shading.” <i>International Journal of Computer Vision</i>. Springer Nature, 2020. <a href=\"https://doi.org/10.1007/s11263-019-01219-8\">https://doi.org/10.1007/s11263-019-01219-8</a>.","ama":"Henderson PM, Ferrari V. Learning single-image 3D reconstruction by generative modelling of shape, pose and shading. <i>International Journal of Computer Vision</i>. 2020;128:835-854. doi:<a href=\"https://doi.org/10.1007/s11263-019-01219-8\">10.1007/s11263-019-01219-8</a>","ieee":"P. M. Henderson and V. Ferrari, “Learning single-image 3D reconstruction by generative modelling of shape, pose and shading,” <i>International Journal of Computer Vision</i>, vol. 128. Springer Nature, pp. 835–854, 2020.","ista":"Henderson PM, Ferrari V. 2020. Learning single-image 3D reconstruction by generative modelling of shape, pose and shading. International Journal of Computer Vision. 128, 835–854."},"date_published":"2020-04-01T00:00:00Z","oa":1,"acknowledgement":"Open access funding provided by Institute of Science and Technology (IST Austria).","intvolume":"       128","article_type":"original","publication_identifier":{"issn":["0920-5691"],"eissn":["1573-1405"]},"volume":128},{"ec_funded":1,"year":"2020","project":[{"_id":"268A44D6-B435-11E9-9278-68D0E5697425","grant_number":"805223","call_identifier":"H2020","name":"Elastic Coordination for Scalable Machine Learning"}],"type":"conference","corr_author":"1","month":"07","conference":{"start_date":"2020-07-12","name":"ICML: International Conference on Machine Learning","end_date":"2020-07-18","location":"Online"},"publisher":"ML Research Press","day":"12","related_material":{"link":[{"relation":"supplementary_material","url":"http://proceedings.mlr.press/v119/konstantinov20a/konstantinov20a-supp.pdf"}],"record":[{"relation":"dissertation_contains","status":"public","id":"10799"}]},"language":[{"iso":"eng"}],"ddc":["000"],"date_updated":"2026-04-07T14:19:48Z","file":[{"file_name":"2020_PMLR_Konstantinov.pdf","date_created":"2021-02-15T09:00:01Z","file_id":"9120","checksum":"cc755d0054bc4b2be778ea7aa7884d2f","creator":"dernst","access_level":"open_access","relation":"main_file","content_type":"application/pdf","date_updated":"2021-02-15T09:00:01Z","file_size":281286,"success":1}],"status":"public","external_id":{"arxiv":["2002.10384"]},"publication":"Proceedings of the 37th International Conference on Machine Learning","department":[{"_id":"DaAl"},{"_id":"ChLa"}],"has_accepted_license":"1","article_processing_charge":"No","intvolume":"       119","acknowledgement":"Dan Alistarh is supported in part by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 805223 ScaleML). This research was supported by the Scientific Service Units (SSU) of IST Austria through resources provided by Scientific Computing (SciComp).","publication_identifier":{"issn":["2640-3498"]},"volume":119,"_id":"8724","scopus_import":"1","date_created":"2020-11-05T15:25:58Z","author":[{"full_name":"Konstantinov, Nikola H","id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87","first_name":"Nikola H","last_name":"Konstantinov","orcid":"0009-0009-5204-7621"},{"first_name":"Elias","last_name":"Frantar","full_name":"Frantar, Elias","id":"09a8f98d-ec99-11ea-ae11-c063a7b7fe5f"},{"first_name":"Dan-Adrian","last_name":"Alistarh","orcid":"0000-0003-3650-940X","full_name":"Alistarh, Dan-Adrian","id":"4A899BFC-F248-11E8-B48F-1D18A9856A87"},{"first_name":"Christoph","orcid":"0000-0001-8622-7887","last_name":"Lampert","full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","date_published":"2020-07-12T00:00:00Z","oa":1,"arxiv":1,"citation":{"short":"N.H. Konstantinov, E. Frantar, D.-A. Alistarh, C. Lampert, in:, Proceedings of the 37th International Conference on Machine Learning, ML Research Press, 2020, pp. 5416–5425.","mla":"Konstantinov, Nikola H., et al. “On the Sample Complexity of Adversarial Multi-Source PAC Learning.” <i>Proceedings of the 37th International Conference on Machine Learning</i>, vol. 119, ML Research Press, 2020, pp. 5416–25.","chicago":"Konstantinov, Nikola H, Elias Frantar, Dan-Adrian Alistarh, and Christoph Lampert. “On the Sample Complexity of Adversarial Multi-Source PAC Learning.” In <i>Proceedings of the 37th International Conference on Machine Learning</i>, 119:5416–25. ML Research Press, 2020.","ama":"Konstantinov NH, Frantar E, Alistarh D-A, Lampert C. On the sample complexity of adversarial multi-source PAC learning. In: <i>Proceedings of the 37th International Conference on Machine Learning</i>. Vol 119. ML Research Press; 2020:5416-5425.","apa":"Konstantinov, N. H., Frantar, E., Alistarh, D.-A., &#38; Lampert, C. (2020). On the sample complexity of adversarial multi-source PAC learning. In <i>Proceedings of the 37th International Conference on Machine Learning</i> (Vol. 119, pp. 5416–5425). Online: ML Research Press.","ista":"Konstantinov NH, Frantar E, Alistarh D-A, Lampert C. 2020. On the sample complexity of adversarial multi-source PAC learning. Proceedings of the 37th International Conference on Machine Learning. ICML: International Conference on Machine Learning vol. 119, 5416–5425.","ieee":"N. H. Konstantinov, E. Frantar, D.-A. Alistarh, and C. Lampert, “On the sample complexity of adversarial multi-source PAC learning,” in <i>Proceedings of the 37th International Conference on Machine Learning</i>, Online, 2020, vol. 119, pp. 5416–5425."},"abstract":[{"text":"We study the problem of learning from multiple untrusted data sources, a scenario of increasing practical relevance given the recent emergence of crowdsourcing and collaborative learning paradigms. Specifically, we analyze the situation in which a learning system obtains datasets from multiple sources, some of which might be biased or even adversarially perturbed. It is\r\nknown that in the single-source case, an adversary with the power to corrupt a fixed fraction of the training data can prevent PAC-learnability, that is, even in the limit of infinitely much training data, no learning system can approach the optimal test error. In this work we show that, surprisingly, the same is not true in the multi-source setting, where the adversary can arbitrarily\r\ncorrupt a fixed fraction of the data sources. Our main results are a generalization bound that provides finite-sample guarantees for this learning setting, as well as corresponding lower bounds. Besides establishing PAC-learnability our results also show that in a cooperative learning setting sharing data with other parties has provable benefits, even if some\r\nparticipants are malicious. ","lang":"eng"}],"acknowledged_ssus":[{"_id":"ScienComp"}],"title":"On the sample complexity of adversarial multi-source PAC learning","oa_version":"Published Version","publication_status":"published","page":"5416-5425","quality_controlled":"1","file_date_updated":"2021-02-15T09:00:01Z"},{"author":[{"first_name":"Phuong","last_name":"Bui Thi Mai","full_name":"Bui Thi Mai, Phuong","id":"3EC6EE64-F248-11E8-B48F-1D18A9856A87"},{"last_name":"Lampert","orcid":"0000-0001-8622-7887","first_name":"Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"conference":{"end_date":"2020-04-30","name":"ICLR: International Conference on Learning Representations","start_date":"2020-04-27","location":"Online"},"date_created":"2020-02-11T09:07:37Z","_id":"7481","corr_author":"1","month":"04","oa":1,"date_published":"2020-04-26T00:00:00Z","citation":{"ista":"Phuong M, Lampert C. 2020. Functional vs. parametric equivalence of ReLU networks. 8th International Conference on Learning Representations. ICLR: International Conference on Learning Representations.","ieee":"M. Phuong and C. Lampert, “Functional vs. parametric equivalence of ReLU networks,” in <i>8th International Conference on Learning Representations</i>, Online, 2020.","chicago":"Phuong, Mary, and Christoph Lampert. “Functional vs. Parametric Equivalence of ReLU Networks.” In <i>8th International Conference on Learning Representations</i>, 2020.","mla":"Phuong, Mary, and Christoph Lampert. “Functional vs. Parametric Equivalence of ReLU Networks.” <i>8th International Conference on Learning Representations</i>, 2020.","short":"M. Phuong, C. Lampert, in:, 8th International Conference on Learning Representations, 2020.","ama":"Phuong M, Lampert C. Functional vs. parametric equivalence of ReLU networks. In: <i>8th International Conference on Learning Representations</i>. ; 2020.","apa":"Phuong, M., &#38; Lampert, C. (2020). Functional vs. parametric equivalence of ReLU networks. In <i>8th International Conference on Learning Representations</i>. Online."},"day":"26","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","type":"conference","year":"2020","status":"public","article_processing_charge":"No","publication":"8th International Conference on Learning Representations","has_accepted_license":"1","quality_controlled":"1","file_date_updated":"2020-07-14T12:47:59Z","department":[{"_id":"ChLa"}],"related_material":{"record":[{"status":"public","id":"9418","relation":"dissertation_contains"}],"link":[{"url":"https://iclr.cc/virtual_2020/poster_Bylx-TNKvH.html","relation":"supplementary_material"}]},"language":[{"iso":"eng"}],"abstract":[{"text":"We address the following question:  How redundant is the parameterisation of ReLU networks? Specifically, we consider transformations of the weight space which leave the function implemented by the network intact.  Two such transformations are known for feed-forward architectures:  permutation of neurons within a layer, and positive scaling of all incoming weights of a neuron coupled with inverse scaling of its outgoing weights. In this work, we show for architectures with non-increasing widths that permutation and scaling are in fact the only function-preserving weight transformations.  For any eligible architecture we give an explicit construction of a neural network such that any other network that implements the same function can be obtained from the original one by the application of permutations and rescaling.  The proof relies on a geometric understanding of boundaries between linear regions of ReLU networks, and we hope the developed mathematical tools are of independent interest.","lang":"eng"}],"date_updated":"2026-04-08T07:01:16Z","publication_status":"published","file":[{"access_level":"open_access","creator":"bphuong","file_id":"7482","checksum":"8d372ea5defd8cb8fdc430111ed754a9","date_created":"2020-02-11T09:07:27Z","file_name":"main.pdf","relation":"main_file","content_type":"application/pdf","file_size":405469,"date_updated":"2020-07-14T12:47:59Z"}],"ddc":["000"],"title":"Functional vs. parametric equivalence of ReLU networks","oa_version":"Published Version"}]