{"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/training-deep-learning-models-with-norm","title":"Training Deep Learning Models with Norm-Constrained LMOs","arxiv_id":"2502.07529","date":"2025-02-11","proceeding":null,"authors":["Thomas Pethick","Wanyun Xie","Kimon Antonakopoulos","Zhenyu Zhu","Antonio Silveti-Falls","Volkan Cevher"],"abstract":"In this work, we study optimization methods that leverage the linear minimization oracle (LMO) over a norm-ball. We propose a new stochastic family of algorithms that uses the LMO to adapt to the geometry of the problem and, perhaps surprisingly, show that they can be applied to unconstrained problems. The resulting update rule unifies several existing optimization methods under a single framework. Furthermore, we propose an explicit choice of norm for deep architectures, which, as a side benefit, leads to the transferability of hyperparameters across model sizes. Experimentally, we demonstrate significant speedups on nanoGPT training using our algorithm, Scion, without any reliance on Adam. The proposed method is memory-efficient, requiring only one set of model weights and one set of gradients, which can be stored in half-precision. The code is available at https://github.com/LIONS-EPFL/scion .","url_abs":"https://arxiv.org/abs/2502.07529v2","url_pdf":"https://arxiv.org/pdf/2502.07529v2.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":[{"paper_slug":"training-deep-learning-models-with-norm","repo_url":"https://github.com/lions-epfl/scion","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"training-deep-learning-models-with-norm","repo_url":"https://github.com/LunNova/ScionLight-reimpl","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"set","method_name":"SET"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2502.07529","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2502.07529"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/lions-epfl/scion","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/LunNova/ScionLight-reimpl","reach":null}],"summary":{"ran_honours":1,"unverified":2},"by_repo_kind":{"official":{"samples":2,"ran":1,"repositories":1},"listed":{"samples":1,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"9112c632b6c7c298","entry":"zeroth_power_via_svd","repo":"lions-epfl/scion","repo_kind":"official","path":"scion.py","file_url":"https://github.com/lions-epfl/scion/blob/HEAD/scion.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9112c632b6c7c298"}},{"code_sha256_prefix":"ebf4909ddc522784","entry":"zeropower_via_newtonschulz5","repo":"lions-epfl/scion","repo_kind":"official","path":"scion.py","file_url":"https://github.com/lions-epfl/scion/blob/HEAD/scion.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ebf4909ddc522784"}},{"code_sha256_prefix":"242fd0f7cbb63b14","entry":"zeropower_via_newtonschulz5","repo":"LunNova/ScionLight-reimpl","repo_kind":"listed","path":"scionlight.py","file_url":"https://github.com/LunNova/ScionLight-reimpl/blob/HEAD/scionlight.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"242fd0f7cbb63b14"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}