{"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/geometry-aware-gradient-algorithms-for-neural","title":"Geometry-Aware Gradient Algorithms for Neural Architecture Search","arxiv_id":"2004.07802","date":"2020-04-16","proceeding":"ICLR 2021 1","authors":["Liam Li","Mikhail Khodak","Maria-Florina Balcan","Ameet Talwalkar"],"abstract":"Recent state-of-the-art methods for neural architecture search (NAS) exploit gradient-based optimization by relaxing the problem into continuous optimization over architectures and shared-weights, a noisy process that remains poorly understood. We argue for the study of single-level empirical risk minimization to understand NAS with weight-sharing, reducing the design of NAS methods to devising optimizers and regularizers that can quickly obtain high-quality solutions to this problem. Invoking the theory of mirror descent, we present a geometry-aware framework that exploits the underlying structure of this optimization to return sparse architectural parameters, leading to simple yet novel algorithms that enjoy fast convergence guarantees and achieve state-of-the-art accuracy on the latest NAS benchmarks in computer vision. Notably, we exceed the best published results for both CIFAR and ImageNet on both the DARTS search space and NAS-Bench201; on the latter we achieve near-oracle-optimal performance on CIFAR-10 and CIFAR-100. Together, our theory and experiments demonstrate a principled way to co-design optimizers and continuous relaxations of discrete NAS search spaces.","url_abs":"https://arxiv.org/abs/2004.07802v5","url_pdf":"https://arxiv.org/pdf/2004.07802v5.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":"geometry-aware-gradient-algorithms-for-neural","repo_url":"https://github.com/liamcli/gaea_release","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"architecture-search","task_name":"Neural Architecture Search"}],"methods":[{"method_slug":"darts","method_name":"DARTS"},{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/neural-architecture-search-on-imagenet","task":"Neural Architecture Search","dataset":"ImageNet","model":"GAEA PC-DARTS","rank_in_archive_order":97,"of":135,"metrics":{"Params":"5.6","Top-1 Error Rate":"24"},"uses_additional_data":false},{"leaderboard":"/sota/neural-architecture-search-on-nas-bench-201-1","task":"Neural Architecture Search","dataset":"NAS-Bench-201, CIFAR-10","model":"GAEA DARTS (ERM)","rank_in_archive_order":16,"of":37,"metrics":{"Accuracy (Test)":"94.1"},"uses_additional_data":false},{"leaderboard":"/sota/neural-architecture-search-on-nas-bench-201-2","task":"Neural Architecture Search","dataset":"NAS-Bench-201, CIFAR-100","model":"GAEA DARTS (ERM)","rank_in_archive_order":9,"of":40,"metrics":{"Accuracy (Test)":"73.43"},"uses_additional_data":false},{"leaderboard":"/sota/neural-architecture-search-on-nas-bench-201","task":"Neural Architecture Search","dataset":"NAS-Bench-201, ImageNet-16-120","model":"GAEA DARTS (ERM)","rank_in_archive_order":10,"of":49,"metrics":{"Accuracy (Test)":"46.36"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2004.07802","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2004.07802"}},"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":"deterministic:regex_extraction","url":"https://github.com/automl/nasbench-1shot1","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/NivNayman/XNAS","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/liamcli/gaea_release","reach":null}],"summary":{"ran":10,"ran_honours":1,"ran_draft_wrong":1,"unverified":3},"by_repo_kind":{"official":{"samples":5,"ran":3,"repositories":1},"found_in_text":{"samples":10,"ran":9,"repositories":2}},"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":5,"samples":[{"code_sha256_prefix":"0ae6d8b3f048531d","entry":"AdaptiveLR","repo":"liamcli/gaea_release","repo_kind":"official","path":"cnn/architect/architect_edarts.py","file_url":"https://github.com/liamcli/gaea_release/blob/HEAD/cnn/architect/architect_edarts.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0ae6d8b3f048531d"}},{"code_sha256_prefix":"b0d6e9e65ecc665d","entry":"Architect","repo":"automl/nasbench-1shot1","repo_kind":"found_in_text","path":"optimizers/darts/architect.py","file_url":"https://github.com/automl/nasbench-1shot1/blob/HEAD/optimizers/darts/architect.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b0d6e9e65ecc665d"}},{"code_sha256_prefix":"132f546bb7ef86ba","entry":"DilConv","repo":"NivNayman/XNAS","repo_kind":"found_in_text","path":"model.py","file_url":"https://github.com/NivNayman/XNAS/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"132f546bb7ef86ba"}},{"code_sha256_prefix":"5acfdb2e37fd4c79","entry":"FactorizedReduce","repo":"NivNayman/XNAS","repo_kind":"found_in_text","path":"model.py","file_url":"https://github.com/NivNayman/XNAS/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"5acfdb2e37fd4c79"}},{"code_sha256_prefix":"2b798f4102f93552","entry":"History","repo":"liamcli/gaea_release","repo_kind":"official","path":"cnn/architect/architect_edarts.py","file_url":"https://github.com/liamcli/gaea_release/blob/HEAD/cnn/architect/architect_edarts.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"2b798f4102f93552"}},{"code_sha256_prefix":"c6c6288219c9d532","entry":"Identity","repo":"NivNayman/XNAS","repo_kind":"found_in_text","path":"model.py","file_url":"https://github.com/NivNayman/XNAS/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c6c6288219c9d532"}},{"code_sha256_prefix":"2b614488c8cb14d3","entry":"ReLUConvBN","repo":"NivNayman/XNAS","repo_kind":"found_in_text","path":"model.py","file_url":"https://github.com/NivNayman/XNAS/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"2b614488c8cb14d3"}},{"code_sha256_prefix":"290cc06f55a8d0ed","entry":"SepConv","repo":"NivNayman/XNAS","repo_kind":"found_in_text","path":"model.py","file_url":"https://github.com/NivNayman/XNAS/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"290cc06f55a8d0ed"}},{"code_sha256_prefix":"c585344417dc92cc","entry":"SepConvRes","repo":"NivNayman/XNAS","repo_kind":"found_in_text","path":"model.py","file_url":"https://github.com/NivNayman/XNAS/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c585344417dc92cc"}},{"code_sha256_prefix":"bc41443a9046202e","entry":"Zero","repo":"NivNayman/XNAS","repo_kind":"found_in_text","path":"model.py","file_url":"https://github.com/NivNayman/XNAS/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"bc41443a9046202e"}},{"code_sha256_prefix":"acb824e0877d8cbd","entry":"_concat","repo":"automl/nasbench-1shot1","repo_kind":"found_in_text","path":"optimizers/darts/architect.py","file_url":"https://github.com/automl/nasbench-1shot1/blob/HEAD/optimizers/darts/architect.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"acb824e0877d8cbd"}},{"code_sha256_prefix":"35516df482cccb4b","entry":"normalize","repo":"liamcli/gaea_release","repo_kind":"official","path":"cnn/architect/architect_edarts.py","file_url":"https://github.com/liamcli/gaea_release/blob/HEAD/cnn/architect/architect_edarts.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":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"35516df482cccb4b"}},{"code_sha256_prefix":"e70f49be6527cff9","entry":"Architect","repo":"liamcli/gaea_release","repo_kind":"official","path":"cnn/architect/architect_edarts.py","file_url":"https://github.com/liamcli/gaea_release/blob/HEAD/cnn/architect/architect_edarts.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"e70f49be6527cff9"}},{"code_sha256_prefix":"411c70cf0de59d37","entry":"ArchitectEDARTS","repo":"liamcli/gaea_release","repo_kind":"official","path":"cnn/architect/architect_edarts.py","file_url":"https://github.com/liamcli/gaea_release/blob/HEAD/cnn/architect/architect_edarts.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"411c70cf0de59d37"}},{"code_sha256_prefix":"f4c89acbb418f526","entry":"Cell","repo":"NivNayman/XNAS","repo_kind":"found_in_text","path":"model.py","file_url":"https://github.com/NivNayman/XNAS/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f4c89acbb418f526"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}