{"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/adam-exploits-ell-infty-geometry-of-loss","title":"Adam Exploits $\\ell_\\infty$-geometry of Loss Landscape via Coordinate-wise Adaptivity","arxiv_id":"2410.08198","date":"2024-10-10","proceeding":null,"authors":["Shuo Xie","Mohamad Amin Mohamadi","Zhiyuan Li"],"abstract":"Adam outperforms SGD when training language models. Yet this advantage is not well-understood theoretically -- previous convergence analysis for Adam and SGD mainly focuses on the number of steps $T$ and is already minimax-optimal in non-convex cases, which are both $\\widetilde{O}(T^{-1/4})$. In this work, we argue that the exploitation of nice $\\ell_\\infty$-geometry is the key advantage of Adam over SGD. More specifically, we give a new convergence analysis for Adam under novel assumptions that loss is smooth under $\\ell_\\infty$-geometry rather than the more common $\\ell_2$-geometry, which yields a much better empirical smoothness constant for GPT-2 and ResNet models. Our experiments confirm that Adam performs much worse when the favorable $\\ell_\\infty$-geometry is changed while SGD provably remains unaffected. We also extend the convergence analysis to blockwise Adam under novel blockwise smoothness assumptions.","url_abs":"https://arxiv.org/abs/2410.08198v1","url_pdf":"https://arxiv.org/pdf/2410.08198v1.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":"adam-exploits-ell-infty-geometry-of-loss","repo_url":"https://github.com/mohamad-amin/adam-coordinate-adaptivity","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"jax","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"cosine-annealing","method_name":"Cosine Annealing"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"discriminative-fine-tuning","method_name":"Discriminative Fine-Tuning"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt-2","method_name":"GPT-2"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-cosine-annealing","method_name":"Linear Warmup With Cosine Annealing"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"sgd","method_name":"SGD"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"weight-decay","method_name":"Weight Decay"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2410.08198","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.08198"}},"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/mohamad-amin/adam-coordinate-adaptivity","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":6,"ran_draft_wrong":1,"unverified":3},"by_repo_kind":{"official":{"samples":10,"ran":7,"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":"fbf63052382288a7","entry":"UID","repo":"mohamad-amin/adam-coordinate-adaptivity","repo_kind":"official","path":"src/util/util.py","file_url":"https://github.com/mohamad-amin/adam-coordinate-adaptivity/blob/HEAD/src/util/util.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fbf63052382288a7"}},{"code_sha256_prefix":"5b4861bd8def8524","entry":"get_or_make","repo":"mohamad-amin/adam-coordinate-adaptivity","repo_kind":"official","path":"src/util/jax_utils.py","file_url":"https://github.com/mohamad-amin/adam-coordinate-adaptivity/blob/HEAD/src/util/jax_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5b4861bd8def8524"}},{"code_sha256_prefix":"c376a3d8fa47fdfc","entry":"load_log","repo":"mohamad-amin/adam-coordinate-adaptivity","repo_kind":"official","path":"src/util/util.py","file_url":"https://github.com/mohamad-amin/adam-coordinate-adaptivity/blob/HEAD/src/util/util.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c376a3d8fa47fdfc"}},{"code_sha256_prefix":"e60ddbbb9cc51133","entry":"map_torch_name_to_flax_key","repo":"mohamad-amin/adam-coordinate-adaptivity","repo_kind":"official","path":"src/util/torch_to_flax.py","file_url":"https://github.com/mohamad-amin/adam-coordinate-adaptivity/blob/HEAD/src/util/torch_to_flax.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e60ddbbb9cc51133"}},{"code_sha256_prefix":"92101b06ecbf1f1e","entry":"new_gelu","repo":"mohamad-amin/adam-coordinate-adaptivity","repo_kind":"official","path":"src/model/model.py","file_url":"https://github.com/mohamad-amin/adam-coordinate-adaptivity/blob/HEAD/src/model/model.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"92101b06ecbf1f1e"}},{"code_sha256_prefix":"22ac4a608be0570c","entry":"tree_load","repo":"mohamad-amin/adam-coordinate-adaptivity","repo_kind":"official","path":"src/util/util.py","file_url":"https://github.com/mohamad-amin/adam-coordinate-adaptivity/blob/HEAD/src/util/util.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"22ac4a608be0570c"}},{"code_sha256_prefix":"2593bc3ed5a143b0","entry":"tree_path_to_name","repo":"mohamad-amin/adam-coordinate-adaptivity","repo_kind":"official","path":"src/util/jax_utils.py","file_url":"https://github.com/mohamad-amin/adam-coordinate-adaptivity/blob/HEAD/src/util/jax_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2593bc3ed5a143b0"}},{"code_sha256_prefix":"7c991ceb6053723f","entry":"create_diagonal_matrix","repo":"mohamad-amin/adam-coordinate-adaptivity","repo_kind":"official","path":"src/engine/hessian/jax_hessian_eval_engine.py","file_url":"https://github.com/mohamad-amin/adam-coordinate-adaptivity/blob/HEAD/src/engine/hessian/jax_hessian_eval_engine.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":"7c991ceb6053723f"}},{"code_sha256_prefix":"206f6bb7e0e53762","entry":"random_split_like_tree","repo":"mohamad-amin/adam-coordinate-adaptivity","repo_kind":"official","path":"src/util/jax_utils.py","file_url":"https://github.com/mohamad-amin/adam-coordinate-adaptivity/blob/HEAD/src/util/jax_utils.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":"206f6bb7e0e53762"}},{"code_sha256_prefix":"03707f1467bf6fb3","entry":"transfer_state_dict_to_params","repo":"mohamad-amin/adam-coordinate-adaptivity","repo_kind":"official","path":"src/util/torch_to_flax.py","file_url":"https://github.com/mohamad-amin/adam-coordinate-adaptivity/blob/HEAD/src/util/torch_to_flax.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":"03707f1467bf6fb3"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}