[{"department":[{"_id":"ChLa"}],"type":"preprint","related_material":{"record":[{"status":"public","relation":"dissertation_contains","id":"21198"}]},"month":"05","corr_author":"1","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","oa_version":"Preprint","date_created":"2026-02-10T08:20:59Z","date_updated":"2026-07-22T06:34:28Z","doi":"10.48550/ARXIV.2505.15579","OA_type":"green","day":"21","tmp":{"short":"CC BY (4.0)","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","image":"/images/cc_by.png","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode"},"citation":{"ama":"Zakerinia H, Scott JA, Lampert C. Federated learning with unlabeled clients: Personalization can happen in low dimensions. <i>arXiv</i>. doi:<a href=\"https://doi.org/10.48550/ARXIV.2505.15579\">10.48550/ARXIV.2505.15579</a>","mla":"Zakerinia, Hossein, et al. “Federated Learning with Unlabeled Clients: Personalization Can Happen in Low Dimensions.” <i>ArXiv</i>, 2505.15579, doi:<a href=\"https://doi.org/10.48550/ARXIV.2505.15579\">10.48550/ARXIV.2505.15579</a>.","apa":"Zakerinia, H., Scott, J. A., &#38; Lampert, C. (n.d.). Federated learning with unlabeled clients: Personalization can happen in low dimensions. <i>arXiv</i>. <a href=\"https://doi.org/10.48550/ARXIV.2505.15579\">https://doi.org/10.48550/ARXIV.2505.15579</a>","ieee":"H. Zakerinia, J. A. Scott, and C. Lampert, “Federated learning with unlabeled clients: Personalization can happen in low dimensions,” <i>arXiv</i>. .","ista":"Zakerinia H, Scott JA, Lampert C. Federated learning with unlabeled clients: Personalization can happen in low dimensions. arXiv, 2505.15579.","chicago":"Zakerinia, Hossein, Jonathan A Scott, and Christoph Lampert. “Federated Learning with Unlabeled Clients: Personalization Can Happen in Low Dimensions.” <i>ArXiv</i>, n.d. <a href=\"https://doi.org/10.48550/ARXIV.2505.15579\">https://doi.org/10.48550/ARXIV.2505.15579</a>.","short":"H. Zakerinia, J.A. Scott, C. Lampert, ArXiv (n.d.)."},"fulldoi":"https://doi.org/10.48550/ARXIV.2505.15579","oa":1,"main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2505.15579"}],"OA_place":"repository","title":"Federated learning with unlabeled clients: Personalization can happen in low dimensions","author":[{"first_name":"Hossein","orcid":"0009-0007-3977-6462","last_name":"Zakerinia","id":"653bd8b6-f394-11eb-9cf6-c0bbf6cd78d4","full_name":"Zakerinia, Hossein"},{"last_name":"Scott","first_name":"Jonathan A","full_name":"Scott, Jonathan A","id":"e499926b-f6e0-11ea-865d-9c63db0031e8"},{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887"}],"date_published":"2025-05-21T00:00:00Z","publication_status":"draft","year":"2025","publication":"arXiv","language":[{"iso":"eng"}],"status":"public","abstract":[{"lang":"eng","text":"Personalized federated learning has emerged as a popular approach to training on devices holding statistically heterogeneous data, known as clients. However, most existing approaches require a client to have labeled data for training or finetuning in order to obtain their own personalized model. In this paper we address this by proposing FLowDUP, a novel method that is able to generate a personalized model using only a forward pass with unlabeled data. The generated model parameters reside in a low-dimensional subspace, enabling efficient communication and computation. FLowDUP's learning objective is theoretically motivated by our new transductive multi-task PAC-Bayesian generalization bound, that provides performance guarantees for unlabeled clients. The objective is structured in such a way that it allows both clients with labeled data and clients with only unlabeled data to contribute to the training process. To supplement our theoretical results we carry out a thorough experimental evaluation of FLowDUP, demonstrating strong empirical performance on a range of datasets with differing sorts of statistically heterogeneous clients. Through numerous ablation studies, we test the efficacy of the individual components of the method."}],"article_number":"2505.15579","das_tickbox":"1","_id":"21207","article_processing_charge":"No"},{"year":"2025","publication_status":"published","researchdata_availability":"no","author":[{"orcid":"0009-0007-3977-6462","first_name":"Hossein","last_name":"Zakerinia","id":"653bd8b6-f394-11eb-9cf6-c0bbf6cd78d4","full_name":"Zakerinia, Hossein"},{"last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"title":"Fast rate bounds for multi-task and meta-learning with different sample sizes","date_published":"2025-12-02T00:00:00Z","das_tickbox":"0","quality_controlled":"1","alternative_title":["Advances in Neural Information Processing Systems"],"article_processing_charge":"No","publisher":"Neural Information Processing Systems Foundation","_id":"22824","publication":"39th Conference on Neural Information Processing Systems","volume":38,"acknowledgement":"This research was supported by the Scientific Service Units (SSU) of ISTA through resources provided by Scientific Computing (SciComp).","language":[{"iso":"eng"}],"status":"public","intvolume":"        38","abstract":[{"text":"We present new fast-rate PAC-Bayesian generalization bounds for multi-task and\r\nmeta-learning in the unbalanced setting, i.e. when the tasks have training sets of\r\ndifferent sizes, as is typically the case in real-world scenarios. Previously, only\r\nstandard-rate bounds were known for this situation, while fast-rate bounds were\r\nlimited to the setting where all training sets are of equal size. Our new bounds\r\nare numerically computable as well as interpretable, and we demonstrate their\r\nflexibility in handling a number of cases where they give stronger guarantees\r\nthan previous bounds. Besides the bounds themselves, we also make conceptual\r\ncontributions: we demonstrate that the unbalanced multi-task setting has different\r\nstatistical properties than the balanced situation, specifically that proofs from\r\nthe balanced situation do not carry over to the unbalanced setting. Additionally,\r\nwe shed light on the fact that the unbalanced situation allows two meaningful\r\ndefinitions of multi-task risk, depending on whether all tasks should be considered\r\nequally important or if sample-rich tasks should receive more weight than samplepoor ones.","lang":"eng"}],"conference":{"end_date":"2025-12-07","location":"San Diego, CA, United States","name":"NeurIPS: Neural Information Processing Systems","start_date":"2025-12-02"},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","corr_author":"1","oa_version":"Published Version","month":"12","date_created":"2026-09-06T22:01:59Z","doi":"10.52202/085713-0278","date_updated":"2026-09-10T08:37:33Z","type":"conference","supplementarymaterial":"yes","department":[{"_id":"GradSch"},{"_id":"ChLa"}],"page":"9062-9093","fulldoi":"https://doi.org/10.52202/085713-0278","publication_identifier":{"isbn":["9798331338275"],"issn":["1049-5258"]},"acknowledged_ssus":[{"_id":"ScienComp"}],"citation":{"short":"H. Zakerinia, C. Lampert, in:, 39th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2025, pp. 9062–9093.","ista":"Zakerinia H, Lampert C. 2025. Fast rate bounds for multi-task and meta-learning with different sample sizes. 39th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 38, 9062–9093.","chicago":"Zakerinia, Hossein, and Christoph Lampert. “Fast Rate Bounds for Multi-Task and Meta-Learning with Different Sample Sizes.” In <i>39th Conference on Neural Information Processing Systems</i>, 38:9062–93. Neural Information Processing Systems Foundation, 2025. <a href=\"https://doi.org/10.52202/085713-0278\">https://doi.org/10.52202/085713-0278</a>.","ama":"Zakerinia H, Lampert C. Fast rate bounds for multi-task and meta-learning with different sample sizes. In: <i>39th Conference on Neural Information Processing Systems</i>. Vol 38. Neural Information Processing Systems Foundation; 2025:9062-9093. doi:<a href=\"https://doi.org/10.52202/085713-0278\">10.52202/085713-0278</a>","ieee":"H. Zakerinia and C. Lampert, “Fast rate bounds for multi-task and meta-learning with different sample sizes,” in <i>39th Conference on Neural Information Processing Systems</i>, San Diego, CA, United States, 2025, vol. 38, pp. 9062–9093.","apa":"Zakerinia, H., &#38; Lampert, C. (2025). Fast rate bounds for multi-task and meta-learning with different sample sizes. In <i>39th Conference on Neural Information Processing Systems</i> (Vol. 38, pp. 9062–9093). San Diego, CA, United States: Neural Information Processing Systems Foundation. <a href=\"https://doi.org/10.52202/085713-0278\">https://doi.org/10.52202/085713-0278</a>","mla":"Zakerinia, Hossein, and Christoph Lampert. “Fast Rate Bounds for Multi-Task and Meta-Learning with Different Sample Sizes.” <i>39th Conference on Neural Information Processing Systems</i>, vol. 38, Neural Information Processing Systems Foundation, 2025, pp. 9062–93, doi:<a href=\"https://doi.org/10.52202/085713-0278\">10.52202/085713-0278</a>."},"OA_place":"publisher","scopus_import":"1","oa":1,"main_file_link":[{"open_access":"1","url":"https://doi.org/10.52202/085713-0278"}],"day":"02","OA_type":"free access"},{"page":"3448-3456","department":[{"_id":"DaAl"},{"_id":"ChLa"}],"type":"conference","date_updated":"2024-10-09T21:08:57Z","date_created":"2024-06-02T22:00:57Z","month":"05","oa_version":"Preprint","conference":{"name":"AISTATS: Conference on Artificial Intelligence and Statistics","start_date":"2024-05-02","end_date":"2024-05-04","location":"Valencia, Spain"},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","corr_author":"1","arxiv":1,"day":"01","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2206.10032"}],"oa":1,"scopus_import":"1","publication_identifier":{"eissn":["2640-3498"]},"citation":{"mla":"Zakerinia, Hossein, et al. “Communication-Efficient Federated Learning with Data and Client Heterogeneity.” <i>Proceedings of the 27th International Conference on Artificial Intelligence and Statistics</i>, vol. 238, ML Research Press, 2024, pp. 3448–56.","ieee":"H. Zakerinia, S. Talaei, G. Nadiradze, and D.-A. Alistarh, “Communication-efficient federated learning with data and client heterogeneity,” in <i>Proceedings of the 27th International Conference on Artificial Intelligence and Statistics</i>, Valencia, Spain, 2024, vol. 238, pp. 3448–3456.","apa":"Zakerinia, H., Talaei, S., Nadiradze, G., &#38; Alistarh, D.-A. (2024). Communication-efficient federated learning with data and client heterogeneity. In <i>Proceedings of the 27th International Conference on Artificial Intelligence and Statistics</i> (Vol. 238, pp. 3448–3456). Valencia, Spain: ML Research Press.","ama":"Zakerinia H, Talaei S, Nadiradze G, Alistarh D-A. Communication-efficient federated learning with data and client heterogeneity. In: <i>Proceedings of the 27th International Conference on Artificial Intelligence and Statistics</i>. Vol 238. ML Research Press; 2024:3448-3456.","chicago":"Zakerinia, Hossein, Shayan Talaei, Giorgi Nadiradze, and Dan-Adrian Alistarh. “Communication-Efficient Federated Learning with Data and Client Heterogeneity.” In <i>Proceedings of the 27th International Conference on Artificial Intelligence and Statistics</i>, 238:3448–56. ML Research Press, 2024.","ista":"Zakerinia H, Talaei S, Nadiradze G, Alistarh D-A. 2024. Communication-efficient federated learning with data and client heterogeneity. Proceedings of the 27th International Conference on Artificial Intelligence and Statistics. AISTATS: Conference on Artificial Intelligence and Statistics, PMLR, vol. 238, 3448–3456.","short":"H. Zakerinia, S. Talaei, G. Nadiradze, D.-A. Alistarh, in:, Proceedings of the 27th International Conference on Artificial Intelligence and Statistics, ML Research Press, 2024, pp. 3448–3456."},"date_published":"2024-05-01T00:00:00Z","title":"Communication-efficient federated learning with data and client heterogeneity","author":[{"id":"653bd8b6-f394-11eb-9cf6-c0bbf6cd78d4","full_name":"Zakerinia, Hossein","last_name":"Zakerinia","first_name":"Hossein"},{"full_name":"Talaei, Shayan","first_name":"Shayan","last_name":"Talaei"},{"orcid":"0000-0001-5634-0731","first_name":"Giorgi","last_name":"Nadiradze","id":"3279A00C-F248-11E8-B48F-1D18A9856A87","full_name":"Nadiradze, Giorgi"},{"id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","full_name":"Alistarh, Dan-Adrian","last_name":"Alistarh","first_name":"Dan-Adrian","orcid":"0000-0003-3650-940X"}],"publication_status":"published","year":"2024","abstract":[{"text":"Federated Learning (FL) enables large-scale distributed training of machine learning models, while still allowing individual nodes to maintain data locally. However, executing FL at scale comes with inherent practical challenges: 1) heterogeneity of the local node data distributions, 2) heterogeneity of node computational speeds (asynchrony), but also 3) constraints in the amount of communication between the clients and the server. In this work, we present the first variant of the classic federated averaging (FedAvg) algorithm which, at the same time, supports data heterogeneity, partial client asynchrony, and communication compression. Our algorithm comes with a novel, rigorous analysis showing that, in spite of these system relaxations, it can provide similar convergence to FedAvg in interesting parameter regimes. Experimental results in the rigorous LEAF benchmark on setups of up to 300 nodes show that our algorithm ensures fast convergence for standard federated tasks, improving upon prior quantized and asynchronous approaches.","lang":"eng"}],"intvolume":"       238","status":"public","language":[{"iso":"eng"}],"volume":238,"publication":"Proceedings of the 27th International Conference on Artificial Intelligence and Statistics","_id":"17093","article_processing_charge":"No","publisher":"ML Research Press","external_id":{"arxiv":["2206.10032"]},"quality_controlled":"1","alternative_title":["PMLR"]},{"date_updated":"2026-04-07T11:46:11Z","date_created":"2024-08-11T22:01:12Z","month":"03","corr_author":"1","arxiv":1,"conference":{"location":"Vienna, Austria","end_date":"2024-03-07","name":"ICLR: International Conference on Learning Representations","start_date":"2024-03-07"},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","oa_version":"Published Version","ddc":["000"],"related_material":{"record":[{"id":"21198","relation":"dissertation_contains","status":"public"}]},"department":[{"_id":"ChLa"}],"type":"conference","scopus_import":"1","oa":1,"acknowledged_ssus":[{"_id":"ScienComp"}],"citation":{"ieee":"J. A. Scott, H. Zakerinia, and C. Lampert, “PEFLL: Personalized federated learning by learning to learn,” in <i>12th International Conference on Learning Representations</i>, Vienna, Austria, 2024.","apa":"Scott, J. A., Zakerinia, H., &#38; Lampert, C. (2024). PEFLL: Personalized federated learning by learning to learn. In <i>12th International Conference on Learning Representations</i>. Vienna, Austria: OpenReview.","mla":"Scott, Jonathan A., et al. “PEFLL: Personalized Federated Learning by Learning to Learn.” <i>12th International Conference on Learning Representations</i>, OpenReview, 2024.","ama":"Scott JA, Zakerinia H, Lampert C. PEFLL: Personalized federated learning by learning to learn. In: <i>12th International Conference on Learning Representations</i>. OpenReview; 2024.","chicago":"Scott, Jonathan A, Hossein Zakerinia, and Christoph Lampert. “PEFLL: Personalized Federated Learning by Learning to Learn.” In <i>12th International Conference on Learning Representations</i>. OpenReview, 2024.","ista":"Scott JA, Zakerinia H, Lampert C. 2024. PEFLL: Personalized federated learning by learning to learn. 12th International Conference on Learning Representations. ICLR: International Conference on Learning Representations.","short":"J.A. Scott, H. Zakerinia, C. Lampert, in:, 12th International Conference on Learning Representations, OpenReview, 2024."},"file":[{"checksum":"81b7ea2e667adaf9c7a7b6b376b1f251","content_type":"application/pdf","creator":"dernst","relation":"main_file","date_created":"2024-08-12T07:38:06Z","date_updated":"2024-08-12T07:38:06Z","file_size":1029219,"file_name":"2024_ICLR_Scott.pdf","file_id":"17415","access_level":"open_access","success":1}],"file_date_updated":"2024-08-12T07:38:06Z","day":"07","publication_status":"published","year":"2024","has_accepted_license":"1","date_published":"2024-03-07T00:00:00Z","title":"PEFLL: Personalized federated learning by learning to learn","author":[{"first_name":"Jonathan A","last_name":"Scott","full_name":"Scott, Jonathan A","id":"e499926b-f6e0-11ea-865d-9c63db0031e8"},{"orcid":"0009-0007-3977-6462","first_name":"Hossein","last_name":"Zakerinia","full_name":"Zakerinia, Hossein","id":"653bd8b6-f394-11eb-9cf6-c0bbf6cd78d4"},{"last_name":"Lampert","orcid":"0000-0001-8622-7887","first_name":"Christoph","full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"_id":"17411","publisher":"OpenReview","article_processing_charge":"No","external_id":{"arxiv":["2306.05515"]},"quality_controlled":"1","status":"public","acknowledgement":"This research was supported by the Scientific Service Units (SSU) of ISTA through resources provided by Scientific Computing (SciComp).\r\n","language":[{"iso":"eng"}],"abstract":[{"lang":"eng","text":"We present PeFLL, a new personalized federated learning algorithm that improves\r\nover the state-of-the-art in three aspects: 1) it produces more accurate models,\r\nespecially in the low-data regime, and not only for clients present during its\r\ntraining phase, but also for any that may emerge in the future; 2) it reduces the\r\namount of on-client computation and client-server communication by providing\r\nfuture clients with ready-to-use personalized models that require no additional\r\nfinetuning or optimization; 3) it comes with theoretical guarantees that establish\r\ngeneralization from the observed clients to future ones.\r\nAt the core of PeFLL lies a learning-to-learn approach that jointly trains an\r\nembedding network and a hypernetwork. The embedding network is used to\r\nrepresent clients in a latent descriptor space in a way that reflects their similarity\r\nto each other. The hypernetwork takes as input such descriptors and outputs the\r\nparameters of fully personalized client models. In combination, both networks\r\nconstitute a learning algorithm that achieves state-of-the-art performance in several\r\npersonalized federated learning benchmarks"}],"publication":"12th International Conference on Learning Representations"},{"year":"2024","publication_status":"published","author":[{"id":"653bd8b6-f394-11eb-9cf6-c0bbf6cd78d4","full_name":"Zakerinia, Hossein","last_name":"Zakerinia","first_name":"Hossein","orcid":"0009-0007-3977-6462"},{"last_name":"Behjati","first_name":"Amin","full_name":"Behjati, Amin"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph","last_name":"Lampert","orcid":"0000-0001-8622-7887","first_name":"Christoph"}],"title":"More flexible PAC-Bayesian meta-learning by learning learning algorithms","date_published":"2024-09-01T00:00:00Z","alternative_title":["PMLR"],"quality_controlled":"1","external_id":{"arxiv":["2402.04054"]},"article_processing_charge":"No","publisher":"ML Research Press","_id":"18118","volume":235,"publication":"Proceedings of the 41st International Conference on Machine Learning","abstract":[{"text":"We introduce a new framework for studying meta-learning methods using PAC-Bayesian theory. Its main advantage over previous work is that it allows for more flexibility in how the transfer of knowledge between tasks is realized. For previous approaches, this could only happen indirectly, by means of learning prior distributions over models. In contrast, the new generalization bounds that we prove express the process of meta-learning much more directly as learning the learning algorithm that should be used for future tasks. The flexibility of our framework makes it suitable to analyze a wide range of meta-learning mechanisms and even design new mechanisms. Other than our theoretical contributions we also show empirically that our framework improves the prediction quality in practical meta-learning mechanisms.","lang":"eng"}],"status":"public","intvolume":"       235","language":[{"iso":"eng"}],"ddc":["000"],"oa_version":"Published Version","arxiv":1,"corr_author":"1","conference":{"end_date":"2024-07-27","location":"Vienna, Austria","start_date":"2024-07-21","name":"ICML: International Conference on Machine Learning"},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","month":"09","date_updated":"2026-06-18T18:01:36Z","date_created":"2024-09-22T22:01:45Z","type":"conference","department":[{"_id":"ChLa"}],"page":"58122-58139","publication_identifier":{"eissn":["2640-3498"]},"citation":{"ieee":"H. Zakerinia, A. Behjati, and C. Lampert, “More flexible PAC-Bayesian meta-learning by learning learning algorithms,” in <i>Proceedings of the 41st International Conference on Machine Learning</i>, Vienna, Austria, 2024, vol. 235, pp. 58122–58139.","mla":"Zakerinia, Hossein, et al. “More Flexible PAC-Bayesian Meta-Learning by Learning Learning Algorithms.” <i>Proceedings of the 41st International Conference on Machine Learning</i>, vol. 235, ML Research Press, 2024, pp. 58122–39.","apa":"Zakerinia, H., Behjati, A., &#38; Lampert, C. (2024). More flexible PAC-Bayesian meta-learning by learning learning algorithms. In <i>Proceedings of the 41st International Conference on Machine Learning</i> (Vol. 235, pp. 58122–58139). Vienna, Austria: ML Research Press.","ama":"Zakerinia H, Behjati A, Lampert C. More flexible PAC-Bayesian meta-learning by learning learning algorithms. In: <i>Proceedings of the 41st International Conference on Machine Learning</i>. Vol 235. ML Research Press; 2024:58122-58139.","chicago":"Zakerinia, Hossein, Amin Behjati, and Christoph Lampert. “More Flexible PAC-Bayesian Meta-Learning by Learning Learning Algorithms.” In <i>Proceedings of the 41st International Conference on Machine Learning</i>, 235:58122–39. ML Research Press, 2024.","ista":"Zakerinia H, Behjati A, Lampert C. 2024. More flexible PAC-Bayesian meta-learning by learning learning algorithms. Proceedings of the 41st International Conference on Machine Learning. ICML: International Conference on Machine Learning, PMLR, vol. 235, 58122–58139.","short":"H. Zakerinia, A. Behjati, C. Lampert, in:, Proceedings of the 41st International Conference on Machine Learning, ML Research Press, 2024, pp. 58122–58139."},"main_file_link":[{"url":" https://doi.org/10.48550/arXiv.2402.04054","open_access":"1"}],"scopus_import":"1","oa":1,"day":"01"}]