{"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/neural-architecture-search-on-imagenet-in-1","title":"Neural Architecture Search on ImageNet in Four GPU Hours: A Theoretically Inspired Perspective","arxiv_id":"2102.11535","date":"2021-02-23","proceeding":"ICLR 2021 1","authors":["Wuyang Chen","Xinyu Gong","Zhangyang Wang"],"abstract":"Neural Architecture Search (NAS) has been explosively studied to automate the discovery of top-performer neural networks. Current works require heavy training of supernet or intensive architecture evaluations, thus suffering from heavy resource consumption and often incurring search bias due to truncated training or approximations. Can we select the best neural architectures without involving any training and eliminate a drastic portion of the search cost? We provide an affirmative answer, by proposing a novel framework called training-free neural architecture search (TE-NAS). TE-NAS ranks architectures by analyzing the spectrum of the neural tangent kernel (NTK) and the number of linear regions in the input space. Both are motivated by recent theory advances in deep networks and can be computed without any training and any label. We show that: (1) these two measurements imply the trainability and expressivity of a neural network; (2) they strongly correlate with the network's test accuracy. Further on, we design a pruning-based NAS mechanism to achieve a more flexible and superior trade-off between the trainability and expressivity during the search. In NAS-Bench-201 and DARTS search spaces, TE-NAS completes high-quality search but only costs 0.5 and 4 GPU hours with one 1080Ti on CIFAR-10 and ImageNet, respectively. We hope our work inspires more attempts in bridging the theoretical findings of deep networks and practical impacts in real NAS applications. Code is available at: https://github.com/VITA-Group/TENAS.","url_abs":"https://arxiv.org/abs/2102.11535v4","url_pdf":"https://arxiv.org/pdf/2102.11535v4.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":"neural-architecture-search-on-imagenet-in-1","repo_url":"https://github.com/VITA-Group/TENAS","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"neural-architecture-search-on-imagenet-in-1","repo_url":"https://github.com/idstcv/ZenNAS","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"neural-architecture-search-on-imagenet-in-1","repo_url":"https://github.com/pnnl/torchntk","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"neural-architecture-search-on-imagenet-in-1","repo_url":"https://github.com/vita-group/tegnas","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":null,"task_name":"GPU"},{"task_slug":"architecture-search","task_name":"Neural Architecture Search"}],"methods":[{"method_slug":"darts","method_name":"DARTS"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/neural-architecture-search-on-nats-bench-1","task":"Neural Architecture Search","dataset":"NATS-Bench Topology, CIFAR-10","model":"TE-NAS (Chen et al., 2021)","rank_in_archive_order":4,"of":11,"metrics":{"Test Accuracy":"93.90"},"uses_additional_data":false},{"leaderboard":"/sota/neural-architecture-search-on-nats-bench-2","task":"Neural Architecture Search","dataset":"NATS-Bench Topology, CIFAR-100","model":"TE-NAS (Chen et al., 2021)","rank_in_archive_order":5,"of":11,"metrics":{"Test Accuracy":"71.24"},"uses_additional_data":false},{"leaderboard":"/sota/neural-architecture-search-on-nats-bench","task":"Neural Architecture Search","dataset":"NATS-Bench Topology, ImageNet16-120","model":"TE-NAS (Chen et al., 2021)","rank_in_archive_order":6,"of":11,"metrics":{"Test Accuracy":"42.38"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2102.11535","atlas_url":"https://app.syntology.ai/?focus=2102.11535","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2102.11535"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/VITA-Group/TENAS","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/idstcv/ZenNAS","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/pnnl/torchntk","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/vita-group/tegnas","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":1,"ran_fixture":1,"unverified":2},"by_repo_kind":{"official":{"samples":2,"ran":1,"repositories":1},"listed":{"samples":2,"ran":1,"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":2,"samples":[{"code_sha256_prefix":"5adedaaf29d0048f","entry":"cal_entropy","repo":"VITA-Group/TENAS","repo_kind":"official","path":"lib/models/cell_searchs/search_model_darts.py","file_url":"https://github.com/VITA-Group/TENAS/blob/HEAD/lib/models/cell_searchs/search_model_darts.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":2,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5adedaaf29d0048f"}},{"code_sha256_prefix":"8c9e2a9cf3c22af1","entry":"autograd_ntk","repo":"pnnl/torchntk","repo_kind":"listed","path":"torchntk/autograd/autograd_ntk.py","file_url":"https://github.com/pnnl/torchntk/blob/HEAD/torchntk/autograd/autograd_ntk.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"8c9e2a9cf3c22af1"}},{"code_sha256_prefix":"943299cba82721a7","entry":"get_ntk_n","repo":"idstcv/ZenNAS","repo_kind":"listed","path":"ZeroShotProxy/compute_te_nas_score.py","file_url":"https://github.com/idstcv/ZenNAS/blob/HEAD/ZeroShotProxy/compute_te_nas_score.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":"943299cba82721a7"}},{"code_sha256_prefix":"16d05701e6a4d833","entry":"get_ntk_n","repo":"VITA-Group/TENAS","repo_kind":"official","path":"lib/procedures/ntk.py","file_url":"https://github.com/VITA-Group/TENAS/blob/HEAD/lib/procedures/ntk.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":"16d05701e6a4d833"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}