{"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/composing-global-optimizers-to-reasoning","title":"Composing Global Optimizers to Reasoning Tasks via Algebraic Objects in Neural Nets","arxiv_id":"2410.01779","date":"2024-10-02","proceeding":null,"authors":["Yuandong Tian"],"abstract":"We prove rich algebraic structures of the solution space for 2-layer neural networks with quadratic activation and $L_2$ loss, trained on reasoning tasks in Abelian group (e.g., modular addition). Such a rich structure enables analytical construction of global optimal solutions from partial solutions that only satisfy part of the loss, despite its high nonlinearity. We coin the framework as CoGO (Composing Global Optimizers). Specifically, we show that the weight space over different numbers of hidden nodes of the 2-layer network is equipped with a semi-ring algebraic structure, and the loss function to be optimized consists of monomial potentials, which are ring homomorphism, allowing partial solutions to be composed into global ones by ring addition and multiplication. Our experiments show that around $95\\%$ of the solutions obtained by gradient descent match exactly our theoretical constructions. Although the global optimizers constructed only required a small number of hidden nodes, our analysis on gradient dynamics shows that over-parameterization asymptotically decouples training dynamics and is beneficial. We further show that training dynamics favors simpler solutions under weight decay, and thus high-order global optimizers such as perfect memorization are unfavorable. Code can be found at https://github.com/facebookresearch/luckmatters/tree/yuandong3/ssl/real-dataset.","url_abs":"https://arxiv.org/abs/2410.01779v3","url_pdf":"https://arxiv.org/pdf/2410.01779v3.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":"composing-global-optimizers-to-reasoning","repo_url":"https://github.com/facebookresearch/luckmatters","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"memorization","task_name":"Memorization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2410.01779","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.01779"}},"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/facebookresearch/luckmatters","reach":null}],"summary":{"ran":1,"ran_draft_wrong":2,"ran_fixture":1,"ran_violates":2,"unverified":1},"by_repo_kind":{"official":{"samples":7,"ran":6,"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":7,"samples":[{"code_sha256_prefix":"f4a22fe6c31525cc","entry":"ModularAdditionNN","repo":"facebookresearch/luckmatters","repo_kind":"official","path":"ssl/real-dataset/cogo/modular_addition_simple2.py","file_url":"https://github.com/facebookresearch/luckmatters/blob/HEAD/ssl/real-dataset/cogo/modular_addition_simple2.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"f4a22fe6c31525cc"}},{"code_sha256_prefix":"d3baec476b355c8d","entry":"compute_diag_off_diag_avg","repo":"facebookresearch/luckmatters","repo_kind":"official","path":"ssl/real-dataset/cogo/modular_addition_simple2.py","file_url":"https://github.com/facebookresearch/luckmatters/blob/HEAD/ssl/real-dataset/cogo/modular_addition_simple2.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"d3baec476b355c8d"}},{"code_sha256_prefix":"adde72772ab2aebb","entry":"compute_mps","repo":"facebookresearch/luckmatters","repo_kind":"official","path":"ssl/real-dataset/cogo/draw_dyn_fig3.py","file_url":"https://github.com/facebookresearch/luckmatters/blob/HEAD/ssl/real-dataset/cogo/draw_dyn_fig3.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"adde72772ab2aebb"}},{"code_sha256_prefix":"65dc337d2918e0b7","entry":"construct_bases","repo":"facebookresearch/luckmatters","repo_kind":"official","path":"ssl/real-dataset/cogo/analyze_util.py","file_url":"https://github.com/facebookresearch/luckmatters/blob/HEAD/ssl/real-dataset/cogo/analyze_util.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"65dc337d2918e0b7"}},{"code_sha256_prefix":"397c623e247709b3","entry":"fit_diag_11","repo":"facebookresearch/luckmatters","repo_kind":"official","path":"ssl/real-dataset/cogo/modular_addition_simple2.py","file_url":"https://github.com/facebookresearch/luckmatters/blob/HEAD/ssl/real-dataset/cogo/modular_addition_simple2.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"397c623e247709b3"}},{"code_sha256_prefix":"71713a75d37f00c5","entry":"modular_addition","repo":"facebookresearch/luckmatters","repo_kind":"official","path":"ssl/real-dataset/cogo/modular_addition_simple2.py","file_url":"https://github.com/facebookresearch/luckmatters/blob/HEAD/ssl/real-dataset/cogo/modular_addition_simple2.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"71713a75d37f00c5"}},{"code_sha256_prefix":"a78ac335f662ddcd","entry":"load_model_traj","repo":"facebookresearch/luckmatters","repo_kind":"official","path":"ssl/real-dataset/cogo/analyze_util.py","file_url":"https://github.com/facebookresearch/luckmatters/blob/HEAD/ssl/real-dataset/cogo/analyze_util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"a78ac335f662ddcd"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}