{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/gumbel-matrix-routing-for-flexible-multi-task-1","title":"Flexible Multi-task Networks by Learning Parameter Allocation","arxiv_id":"1910.04915","date":"2019-10-10","proceeding":null,"authors":["Krzysztof Maziarz","Efi Kokiopoulou","Andrea Gesmundo","Luciano Sbaiz","Gabor Bartok","Jesse Berent"],"abstract":"This paper proposes a novel learning method for multi-task applications. Multi-task neural networks can learn to transfer knowledge across different tasks by using parameter sharing. However, sharing parameters between unrelated tasks can hurt performance. To address this issue, we propose a framework to learn fine-grained patterns of parameter sharing. Assuming that the network is composed of several components across layers, our framework uses learned binary variables to allocate components to tasks in order to encourage more parameter sharing between related tasks, and discourage parameter sharing otherwise. The binary allocation variables are learned jointly with the model parameters by standard back-propagation thanks to the Gumbel-Softmax reparametrization method. When applied to the Omniglot benchmark, the proposed method achieves a 17% relative reduction of the error rate compared to state-of-the-art.","url_abs":"https://arxiv.org/abs/1910.04915v2","url_pdf":"https://arxiv.org/pdf/1910.04915v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/multi-task-learning-on-omniglot","task":"Multi-Task Learning","dataset":"OMNIGLOT","model":"Gumbel-Matrix Routing","rank_in_archive_order":1,"of":2,"metrics":{"Average Accuracy":"93.52"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1910.04915","atlas_url":"https://app.syntology.ai/?focus=1910.04915","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}