[{"day":"20","publication":"34th International Conference on Neural Information Processing Systems","external_id":{"arxiv":["2006.15055"]},"citation":{"ama":"Locatello F, Weissenborn D, Unterthiner T, et al. Object-centric learning with slot attention. In: <i>34th International Conference on Neural Information Processing Systems</i>. Vol 33. Neural Information Processing Systems Foundation; 2020:11525-11538.","mla":"Locatello, Francesco, et al. “Object-Centric Learning with Slot Attention.” <i>34th International Conference on Neural Information Processing Systems</i>, vol. 33, Neural Information Processing Systems Foundation, 2020, pp. 11525–38.","chicago":"Locatello, Francesco, Dirk Weissenborn, Thomas Unterthiner, Aravindh Mahendran, Georg Heigold, Jakob Uszkoreit, Alexey Dosovitskiy, and Thomas Kipf. “Object-Centric Learning with Slot Attention.” In <i>34th International Conference on Neural Information Processing Systems</i>, 33:11525–38. Neural Information Processing Systems Foundation, 2020.","ista":"Locatello F, Weissenborn D, Unterthiner T, Mahendran A, Heigold G, Uszkoreit J, Dosovitskiy A, Kipf T. 2020. Object-centric learning with slot attention. 34th International Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 33, 11525–11538.","ieee":"F. Locatello <i>et al.</i>, “Object-centric learning with slot attention,” in <i>34th International Conference on Neural Information Processing Systems</i>, Virtual, 2020, vol. 33, pp. 11525–11538.","apa":"Locatello, F., Weissenborn, D., Unterthiner, T., Mahendran, A., Heigold, G., Uszkoreit, J., … Kipf, T. (2020). Object-centric learning with slot attention. In <i>34th International Conference on Neural Information Processing Systems</i> (Vol. 33, pp. 11525–11538). Virtual: Neural Information Processing Systems Foundation.","short":"F. Locatello, D. Weissenborn, T. Unterthiner, A. Mahendran, G. Heigold, J. Uszkoreit, A. Dosovitskiy, T. Kipf, in:, 34th International Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2020, pp. 11525–11538."},"oa_version":"Preprint","title":"Object-centric learning with slot attention","type":"conference","page":"11525-11538","_id":"14326","publication_status":"published","date_updated":"2025-07-10T11:50:47Z","month":"12","intvolume":"        33","quality_controlled":"1","date_published":"2020-12-20T00:00:00Z","article_processing_charge":"No","extern":"1","status":"public","conference":{"end_date":"2020-12-12","start_date":"2020-12-06","location":"Virtual","name":"NeurIPS: Neural Information Processing Systems"},"alternative_title":["Advances in Neural Information Processing Systems"],"department":[{"_id":"FrLo"}],"author":[{"full_name":"Locatello, Francesco","orcid":"0000-0002-4850-0683","first_name":"Francesco","last_name":"Locatello","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"},{"first_name":"Dirk","last_name":"Weissenborn","full_name":"Weissenborn, Dirk"},{"last_name":"Unterthiner","first_name":"Thomas","full_name":"Unterthiner, Thomas"},{"full_name":"Mahendran, Aravindh","first_name":"Aravindh","last_name":"Mahendran"},{"full_name":"Heigold, Georg","last_name":"Heigold","first_name":"Georg"},{"first_name":"Jakob","last_name":"Uszkoreit","full_name":"Uszkoreit, Jakob"},{"full_name":"Dosovitskiy, Alexey","last_name":"Dosovitskiy","first_name":"Alexey"},{"first_name":"Thomas","last_name":"Kipf","full_name":"Kipf, Thomas"}],"language":[{"iso":"eng"}],"year":"2020","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","publication_identifier":{"eissn":["1049-5258"],"isbn":["9781713829546"]},"volume":33,"arxiv":1,"abstract":[{"lang":"eng","text":"Learning object-centric representations of complex scenes is a promising step towards enabling efficient abstract reasoning from low-level perceptual features. Yet, most deep learning approaches learn distributed representations that do not capture the compositional properties of natural scenes. In this paper, we present the Slot Attention module, an architectural component that interfaces with perceptual representations such as the output of a convolutional neural network and produces a set of task-dependent abstract representations which we call slots. These slots are exchangeable and can bind to any object in the input by specializing through a competitive procedure over multiple rounds of attention. We empirically demonstrate that Slot Attention can extract object-centric representations that enable generalization to unseen compositions when trained on unsupervised object discovery and supervised property prediction tasks.\r\n\r\n"}],"main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2006.15055","open_access":"1"}],"date_created":"2023-09-13T12:03:46Z","publisher":"Neural Information Processing Systems Foundation","oa":1},{"project":[{"_id":"268A44D6-B435-11E9-9278-68D0E5697425","grant_number":"805223","call_identifier":"H2020","name":"Elastic Coordination for Scalable Machine Learning"}],"publisher":"Neural Information Processing Systems Foundation","oa":1,"date_created":"2024-03-06T08:35:58Z","acknowledgement":"The authors would like to thank Blair Bilodeau, David Fleet, Mufan Li, and Jeffrey Negrea for\r\nhelpful discussions. FF was supported by OGS Scholarship. DA and IM were supported the\r\nEuropean Research Council (ERC) under the European Union’s Horizon 2020 research and innovation\r\nprogramme (grant agreement No 805223 ScaleML). DMR was supported by an NSERC Discovery\r\nGrant. ARK was supported by NSERC Postdoctoral Fellowship. Resources used in preparing this research were provided, in part, by the Province of Ontario, the Government of Canada through CIFAR, and companies sponsoring the Vector Institute.","arxiv":1,"abstract":[{"text":"Many communication-efficient variants of SGD use gradient quantization schemes. These schemes are often heuristic and fixed over the course of training. We empirically observe that the statistics of gradients of deep models change during the training. Motivated by this observation, we introduce two adaptive quantization schemes, ALQ and AMQ. In both schemes, processors update their compression schemes in parallel by efficiently computing sufficient statistics of a parametric distribution. We improve the validation accuracy by almost 2% on CIFAR-10 and 1% on ImageNet in challenging low-cost communication setups. Our adaptive methods are also significantly more robust to the choice of hyperparameters.\r\n\r\n","lang":"eng"}],"main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2010.12460"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","publication_identifier":{"isbn":["9781713829546"]},"volume":33,"language":[{"iso":"eng"}],"year":"2020","author":[{"full_name":"Faghri, Fartash ","last_name":"Faghri","first_name":"Fartash "},{"full_name":"Tabrizian, Iman ","last_name":"Tabrizian","first_name":"Iman "},{"first_name":"Ilia","last_name":"Markov","id":"D0CF4148-C985-11E9-8066-0BDEE5697425","full_name":"Markov, Ilia"},{"first_name":"Dan-Adrian","last_name":"Alistarh","id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","full_name":"Alistarh, Dan-Adrian","orcid":"0000-0003-3650-940X"},{"last_name":"Roy","first_name":"Daniel ","full_name":"Roy, Daniel "},{"full_name":"Ramezani-Kebrya, Ali ","last_name":"Ramezani-Kebrya","first_name":"Ali "}],"department":[{"_id":"DaAl"}],"alternative_title":["NeurIPS"],"conference":{"start_date":"2020-12-06","end_date":"2020-12-12","location":"Vancouver, Canada","name":"NeurIPS: Neural Information Processing Systems"},"status":"public","article_processing_charge":"No","ec_funded":1,"date_published":"2020-12-10T00:00:00Z","publication_status":"published","date_updated":"2025-04-14T07:49:16Z","_id":"15086","quality_controlled":"1","intvolume":"        33","month":"12","type":"conference","title":"Adaptive gradient quantization for data-parallel SGD","oa_version":"Preprint","citation":{"ieee":"F. Faghri, I. Tabrizian, I. Markov, D.-A. Alistarh, D. Roy, and A. Ramezani-Kebrya, “Adaptive gradient quantization for data-parallel SGD,” in <i>Advances in Neural Information Processing Systems</i>, Vancouver, Canada, 2020, vol. 33.","apa":"Faghri, F., Tabrizian, I., Markov, I., Alistarh, D.-A., Roy, D., &#38; Ramezani-Kebrya, A. (2020). Adaptive gradient quantization for data-parallel SGD. In <i>Advances in Neural Information Processing Systems</i> (Vol. 33). Vancouver, Canada: Neural Information Processing Systems Foundation.","short":"F. Faghri, I. Tabrizian, I. Markov, D.-A. Alistarh, D. Roy, A. Ramezani-Kebrya, in:, Advances in Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2020.","ista":"Faghri F, Tabrizian I, Markov I, Alistarh D-A, Roy D, Ramezani-Kebrya A. 2020. Adaptive gradient quantization for data-parallel SGD. Advances in Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, NeurIPS, vol. 33.","ama":"Faghri F, Tabrizian I, Markov I, Alistarh D-A, Roy D, Ramezani-Kebrya A. Adaptive gradient quantization for data-parallel SGD. In: <i>Advances in Neural Information Processing Systems</i>. Vol 33. Neural Information Processing Systems Foundation; 2020.","mla":"Faghri, Fartash, et al. “Adaptive Gradient Quantization for Data-Parallel SGD.” <i>Advances in Neural Information Processing Systems</i>, vol. 33, Neural Information Processing Systems Foundation, 2020.","chicago":"Faghri, Fartash , Iman  Tabrizian, Ilia Markov, Dan-Adrian Alistarh, Daniel  Roy, and Ali  Ramezani-Kebrya. “Adaptive Gradient Quantization for Data-Parallel SGD.” In <i>Advances in Neural Information Processing Systems</i>, Vol. 33. Neural Information Processing Systems Foundation, 2020."},"day":"10","publication":"Advances in Neural Information Processing Systems","external_id":{"arxiv":["2010.12460"]}},{"alternative_title":["Advances in Neural Information Processing Systems"],"status":"public","conference":{"start_date":"2020-12-06","end_date":"2020-12-12","location":"Vancouver, Canada","name":"NeurIPS: Neural Information Processing Systems"},"year":"2020","language":[{"iso":"eng"}],"author":[{"first_name":"Paul M","last_name":"Henderson","id":"13C09E74-18D9-11E9-8878-32CFE5697425","full_name":"Henderson, Paul M","orcid":"0000-0002-5198-7445"},{"orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","first_name":"Christoph"}],"department":[{"_id":"ChLa"}],"corr_author":"1","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.","acknowledged_ssus":[{"_id":"ScienComp"}],"arxiv":1,"abstract":[{"lang":"eng","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."}],"main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2007.06705"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","volume":33,"publication_identifier":{"isbn":["9781713829546"]},"publisher":"Neural Information Processing Systems Foundation","oa":1,"date_created":"2020-07-31T16:59:19Z","citation":{"short":"P.M. Henderson, C. Lampert, in:, 34th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2020, pp. 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.","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.","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.","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.","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.","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."},"publication":"34th Conference on Neural Information Processing Systems","day":"07","external_id":{"arxiv":["2007.06705"]},"page":"3106–3117","type":"conference","title":"Unsupervised object-centric video generation and decomposition in 3D","oa_version":"Preprint","date_published":"2020-07-07T00:00:00Z","date_updated":"2025-05-14T11:26:57Z","publication_status":"published","_id":"8188","quality_controlled":"1","month":"07","intvolume":"        33","article_processing_charge":"No"},{"date_created":"2021-07-04T22:01:26Z","oa":1,"publisher":"Neural Information Processing Systems Foundation","project":[{"name":"Elastic Coordination for Scalable Machine Learning","call_identifier":"H2020","grant_number":"805223","_id":"268A44D6-B435-11E9-9278-68D0E5697425"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","volume":33,"publication_identifier":{"isbn":["9781713829546"],"issn":["1049-5258"]},"arxiv":1,"acknowledgement":"We thank Marco Mondelli for discussions related to LDPC decoding, and Giorgi Nadiradze for discussions on analysis of relaxed schedulers. This project has received funding from the European Research Council (ERC) under the European\r\nUnion’s Horizon 2020 research and innovation programme (grant agreement No 805223 ScaleML).","abstract":[{"lang":"eng","text":"The ability to leverage large-scale hardware parallelism has been one of the key enablers of the accelerated recent progress in machine learning. Consequently, there has been considerable effort invested into developing efficient parallel variants of classic machine learning algorithms. However, despite the wealth of knowledge on parallelization, some classic machine learning algorithms often prove hard to parallelize efficiently while maintaining convergence. In this paper, we focus on efficient parallel algorithms for the key machine learning task of inference on graphical models, in particular on the fundamental belief propagation algorithm. We address the challenge of efficiently parallelizing this classic paradigm by showing how to leverage scalable relaxed schedulers in this context. We present an extensive empirical study, showing that our approach outperforms previous parallel belief propagation implementations both in terms of scalability and in terms of wall-clock convergence time, on a range of practical applications."}],"main_file_link":[{"url":"https://proceedings.neurips.cc/paper/2020/hash/fdb2c3bab9d0701c4a050a4d8d782c7f-Abstract.html","open_access":"1"}],"department":[{"_id":"DaAl"}],"corr_author":"1","author":[{"full_name":"Aksenov, Vitaly","last_name":"Aksenov","first_name":"Vitaly"},{"first_name":"Dan-Adrian","id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","last_name":"Alistarh","full_name":"Alistarh, Dan-Adrian","orcid":"0000-0003-3650-940X"},{"full_name":"Korhonen, Janne","first_name":"Janne","id":"C5402D42-15BC-11E9-A202-CA2BE6697425","last_name":"Korhonen"}],"year":"2020","language":[{"iso":"eng"}],"status":"public","conference":{"location":"Vancouver, Canada","end_date":"2020-12-12","start_date":"2020-12-06","name":"NeurIPS: Conference on Neural Information Processing Systems"},"alternative_title":["Advances in Neural Information Processing Systems"],"article_processing_charge":"No","_id":"9631","date_updated":"2025-05-14T11:27:33Z","publication_status":"published","intvolume":"        33","month":"12","quality_controlled":"1","date_published":"2020-12-06T00:00:00Z","ec_funded":1,"oa_version":"Published Version","title":"Scalable belief propagation via relaxed scheduling","type":"conference","page":"22361-22372","scopus_import":"1","day":"06","external_id":{"arxiv":["2002.11505"]},"citation":{"chicago":"Aksenov, Vitaly, Dan-Adrian Alistarh, and Janne Korhonen. “Scalable Belief Propagation via Relaxed Scheduling,” 33:22361–72. Neural Information Processing Systems Foundation, 2020.","mla":"Aksenov, Vitaly, et al. <i>Scalable Belief Propagation via Relaxed Scheduling</i>. Vol. 33, Neural Information Processing Systems Foundation, 2020, pp. 22361–72.","ama":"Aksenov V, Alistarh D-A, Korhonen J. Scalable belief propagation via relaxed scheduling. In: Vol 33. Neural Information Processing Systems Foundation; 2020:22361-22372.","ista":"Aksenov V, Alistarh D-A, Korhonen J. 2020. Scalable belief propagation via relaxed scheduling. NeurIPS: Conference on Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 33, 22361–22372.","apa":"Aksenov, V., Alistarh, D.-A., &#38; Korhonen, J. (2020). Scalable belief propagation via relaxed scheduling (Vol. 33, pp. 22361–22372). Presented at the NeurIPS: Conference on Neural Information Processing Systems, Vancouver, Canada: Neural Information Processing Systems Foundation.","short":"V. Aksenov, D.-A. Alistarh, J. Korhonen, in:, Neural Information Processing Systems Foundation, 2020, pp. 22361–22372.","ieee":"V. Aksenov, D.-A. Alistarh, and J. Korhonen, “Scalable belief propagation via relaxed scheduling,” presented at the NeurIPS: Conference on Neural Information Processing Systems, Vancouver, Canada, 2020, vol. 33, pp. 22361–22372."}},{"abstract":[{"lang":"eng","text":"Second-order information, in the form of Hessian- or Inverse-Hessian-vector products, is a fundamental tool for solving optimization problems. Recently, there has been significant interest in utilizing this information in the context of deep\r\nneural networks; however, relatively little is known about the quality of existing approximations in this context. Our work examines this question, identifies issues with existing approaches, and proposes a method called WoodFisher to compute a faithful and efficient estimate of the inverse Hessian. Our main application is to neural network compression, where we build on the classic Optimal Brain Damage/Surgeon framework. We demonstrate that WoodFisher significantly outperforms popular state-of-the-art methods for oneshot pruning. Further, even when iterative, gradual pruning is allowed, our method results in a gain in test accuracy over the state-of-the-art approaches, for standard image classification datasets such as ImageNet ILSVRC. We examine how our method can be extended to take into account first-order information, as well as\r\nillustrate its ability to automatically set layer-wise pruning thresholds and perform compression in the limited-data regime. The code is available at the following link, https://github.com/IST-DASLab/WoodFisher."}],"main_file_link":[{"url":"https://proceedings.neurips.cc/paper/2020/hash/d1ff1ec86b62cd5f3903ff19c3a326b2-Abstract.html","open_access":"1"}],"arxiv":1,"acknowledgement":"This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 805223 ScaleML). Also, we would like to thank Alexander Shevchenko, Alexandra Peste, and other members of the group for fruitful discussions.","volume":33,"publication_identifier":{"isbn":["9781713829546"],"issn":["1049-5258"]},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","project":[{"name":"Elastic Coordination for Scalable Machine Learning","call_identifier":"H2020","grant_number":"805223","_id":"268A44D6-B435-11E9-9278-68D0E5697425"}],"date_created":"2021-07-04T22:01:26Z","oa":1,"publisher":"Neural Information Processing Systems Foundation","conference":{"name":"NeurIPS: Conference on Neural Information Processing Systems","start_date":"2020-12-06","end_date":"2020-12-12","location":"Vancouver, Canada"},"alternative_title":["Advances in Neural Information Processing Systems"],"status":"public","author":[{"full_name":"Singh, Sidak Pal","first_name":"Sidak Pal","id":"DD138E24-D89D-11E9-9DC0-DEF6E5697425","last_name":"Singh"},{"full_name":"Alistarh, Dan-Adrian","orcid":"0000-0003-3650-940X","first_name":"Dan-Adrian","last_name":"Alistarh","id":"4A899BFC-F248-11E8-B48F-1D18A9856A87"}],"year":"2020","language":[{"iso":"eng"}],"corr_author":"1","department":[{"_id":"DaAl"},{"_id":"ToHe"}],"date_published":"2020-12-06T00:00:00Z","ec_funded":1,"intvolume":"        33","month":"12","quality_controlled":"1","_id":"9632","publication_status":"published","date_updated":"2025-05-14T11:27:23Z","article_processing_charge":"No","citation":{"ieee":"S. P. Singh and D.-A. Alistarh, “WoodFisher: Efficient second-order approximation for neural network compression,” presented at the NeurIPS: Conference on Neural Information Processing Systems, Vancouver, Canada, 2020, vol. 33, pp. 18098–18109.","short":"S.P. Singh, D.-A. Alistarh, in:, Neural Information Processing Systems Foundation, 2020, pp. 18098–18109.","apa":"Singh, S. P., &#38; Alistarh, D.-A. (2020). WoodFisher: Efficient second-order approximation for neural network compression (Vol. 33, pp. 18098–18109). Presented at the NeurIPS: Conference on Neural Information Processing Systems, Vancouver, Canada: Neural Information Processing Systems Foundation.","ista":"Singh SP, Alistarh D-A. 2020. WoodFisher: Efficient second-order approximation for neural network compression. NeurIPS: Conference on Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 33, 18098–18109.","mla":"Singh, Sidak Pal, and Dan-Adrian Alistarh. <i>WoodFisher: Efficient Second-Order Approximation for Neural Network Compression</i>. Vol. 33, Neural Information Processing Systems Foundation, 2020, pp. 18098–109.","ama":"Singh SP, Alistarh D-A. WoodFisher: Efficient second-order approximation for neural network compression. In: Vol 33. Neural Information Processing Systems Foundation; 2020:18098-18109.","chicago":"Singh, Sidak Pal, and Dan-Adrian Alistarh. “WoodFisher: Efficient Second-Order Approximation for Neural Network Compression,” 33:18098–109. Neural Information Processing Systems Foundation, 2020."},"external_id":{"arxiv":["2004.14340"]},"day":"06","scopus_import":"1","title":"WoodFisher: Efficient second-order approximation for neural network compression","type":"conference","page":"18098-18109","oa_version":"Published Version"}]