{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2016:BHOWV3D6AQAVD2526IS5OIOHPV","short_pith_number":"pith:BHOWV3D6","schema_version":"1.0","canonical_sha256":"09dd6aec7e040151ebbaf225d721c77d75cdc1636f0234d10916745b722d52ce","source":{"kind":"arxiv","id":"1611.01578","version":2},"attestation_state":"computed","paper":{"title":"Neural Architecture Search with Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.NE"],"primary_cat":"cs.LG","authors_text":"Barret Zoph, Quoc V. Le","submitted_at":"2016-11-05T00:41:37Z","abstract_excerpt":"Neural networks are powerful and flexible models that work well for many difficult learning tasks in image, speech and natural language understanding. Despite their success, neural networks are still hard to design. In this paper, we use a recurrent network to generate the model descriptions of neural networks and train this RNN with reinforcement learning to maximize the expected accuracy of the generated architectures on a validation set. On the CIFAR-10 dataset, our method, starting from scratch, can design a novel network architecture that rivals the best human-invented architecture in ter"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1611.01578","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-11-05T00:41:37Z","cross_cats_sorted":["cs.AI","cs.NE"],"title_canon_sha256":"9d648c1b3ad6bd59423ea0fca91598684cf9101473fe7ab3cacfb8c4c475157a","abstract_canon_sha256":"38e92b38fb05a04fc848bbaed5304c60c63a467143233ee13306b23c1ee6838f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:50:42.333661Z","signature_b64":"HYsDS+cYKxcLjxMFdmNJFBkc2M5ESaH2ChzZj5nw6Pk7+dYTwBGEc2Gl4lLqYTfSaaxYxNKt63mej2IhyA/aAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"09dd6aec7e040151ebbaf225d721c77d75cdc1636f0234d10916745b722d52ce","last_reissued_at":"2026-05-18T00:50:42.332950Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:50:42.332950Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neural Architecture Search with Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.NE"],"primary_cat":"cs.LG","authors_text":"Barret Zoph, Quoc V. Le","submitted_at":"2016-11-05T00:41:37Z","abstract_excerpt":"Neural networks are powerful and flexible models that work well for many difficult learning tasks in image, speech and natural language understanding. Despite their success, neural networks are still hard to design. In this paper, we use a recurrent network to generate the model descriptions of neural networks and train this RNN with reinforcement learning to maximize the expected accuracy of the generated architectures on a validation set. On the CIFAR-10 dataset, our method, starting from scratch, can design a novel network architecture that rivals the best human-invented architecture in ter"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1611.01578","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"1611.01578","created_at":"2026-05-18T00:50:42.333062+00:00"},{"alias_kind":"arxiv_version","alias_value":"1611.01578v2","created_at":"2026-05-18T00:50:42.333062+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1611.01578","created_at":"2026-05-18T00:50:42.333062+00:00"},{"alias_kind":"pith_short_12","alias_value":"BHOWV3D6AQAV","created_at":"2026-05-18T12:30:07.202191+00:00"},{"alias_kind":"pith_short_16","alias_value":"BHOWV3D6AQAVD252","created_at":"2026-05-18T12:30:07.202191+00:00"},{"alias_kind":"pith_short_8","alias_value":"BHOWV3D6","created_at":"2026-05-18T12:30:07.202191+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":45,"internal_anchor_count":34,"sample":[{"citing_arxiv_id":"1906.11080","citing_title":"AGAN: Towards Automated Design of Generative Adversarial Networks","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"1906.11527","citing_title":"Hyp-RL : Hyperparameter Optimization by Reinforcement Learning","ref_index":42,"is_internal_anchor":true},{"citing_arxiv_id":"1906.12061","citing_title":"Learning to Cope with Adversarial Attacks","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"1906.12087","citing_title":"ARMIN: Towards a More Efficient and Light-weight Recurrent Memory Network","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"1906.12348","citing_title":"MLFriend: Interactive Prediction Task Recommendation for Event-Driven Time-Series Data","ref_index":25,"is_internal_anchor":true},{"citing_arxiv_id":"1907.00274","citing_title":"NetTailor: Tuning the Architecture, Not Just the Weights","ref_index":75,"is_internal_anchor":true},{"citing_arxiv_id":"1907.01939","citing_title":"Neural Network Architecture Search with Differentiable Cartesian Genetic Programming for Regression","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"1907.03389","citing_title":"Blending-target Domain Adaptation by Adversarial Meta-Adaptation Networks","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"1907.02871","citing_title":"Genetic Network Architecture Search","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"1907.04648","citing_title":"EPNAS: Efficient Progressive Neural Architecture Search","ref_index":52,"is_internal_anchor":true},{"citing_arxiv_id":"1907.04632","citing_title":"Video Action Recognition Via Neural Architecture Searching","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"1907.05415","citing_title":"Learning to learn with quantum neural networks via classical neural networks","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"1907.06968","citing_title":"A Unified Deep Framework for Joint 3D Pose Estimation and Action Recognition from a Single RGB Camera","ref_index":47,"is_internal_anchor":true},{"citing_arxiv_id":"1907.11849","citing_title":"Genetic Deep Learning for Lung Cancer Screening","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"1907.11587","citing_title":"Self-Adaptive 2D-3D Ensemble of Fully Convolutional Networks for Medical Image Segmentation","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"1907.11943","citing_title":"Learnable Parameter Similarity","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2310.02540","citing_title":"Auto-FP: An Experimental Study of Automated Feature Preprocessing for Tabular Data","ref_index":92,"is_internal_anchor":true},{"citing_arxiv_id":"2405.03420","citing_title":"Implantable Adaptive Cells: A Novel Enhancement for Pre-Trained U-Nets in Medical Image Segmentation","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2408.00601","citing_title":"AutoPV: Automatically Design Your Photovoltaic Power Forecasting Model","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2502.00944","citing_title":"Training speedups via batching for geometric learning: an analysis of static and dynamic algorithms","ref_index":41,"is_internal_anchor":true},{"citing_arxiv_id":"2502.07189","citing_title":"Exploring Vision Neural Network Pruning via Screening Methodology","ref_index":60,"is_internal_anchor":true},{"citing_arxiv_id":"2504.18361","citing_title":"COCO-Inpaint: A Benchmark for Detecting and Localizing Inpainting-Based Image Manipulations","ref_index":77,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22343","citing_title":"Sibyl-AutoResearch: Autonomous Research Needs Self-Evolving Trial-and-Error Harnesses, Not Paper Generators","ref_index":46,"is_internal_anchor":true},{"citing_arxiv_id":"2509.26037","citing_title":"CoLLM-NAS: Collaborative Large Language Models for Efficient Knowledge-Guided Neural Architecture Search","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"2605.07379","citing_title":"RELO: Reinforcement Learning to Localize for Visual Object Tracking","ref_index":253,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BHOWV3D6AQAVD2526IS5OIOHPV","json":"https://pith.science/pith/BHOWV3D6AQAVD2526IS5OIOHPV.json","graph_json":"https://pith.science/api/pith-number/BHOWV3D6AQAVD2526IS5OIOHPV/graph.json","events_json":"https://pith.science/api/pith-number/BHOWV3D6AQAVD2526IS5OIOHPV/events.json","paper":"https://pith.science/paper/BHOWV3D6"},"agent_actions":{"view_html":"https://pith.science/pith/BHOWV3D6AQAVD2526IS5OIOHPV","download_json":"https://pith.science/pith/BHOWV3D6AQAVD2526IS5OIOHPV.json","view_paper":"https://pith.science/paper/BHOWV3D6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1611.01578&json=true","fetch_graph":"https://pith.science/api/pith-number/BHOWV3D6AQAVD2526IS5OIOHPV/graph.json","fetch_events":"https://pith.science/api/pith-number/BHOWV3D6AQAVD2526IS5OIOHPV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BHOWV3D6AQAVD2526IS5OIOHPV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BHOWV3D6AQAVD2526IS5OIOHPV/action/storage_attestation","attest_author":"https://pith.science/pith/BHOWV3D6AQAVD2526IS5OIOHPV/action/author_attestation","sign_citation":"https://pith.science/pith/BHOWV3D6AQAVD2526IS5OIOHPV/action/citation_signature","submit_replication":"https://pith.science/pith/BHOWV3D6AQAVD2526IS5OIOHPV/action/replication_record"}},"created_at":"2026-05-18T00:50:42.333062+00:00","updated_at":"2026-05-18T00:50:42.333062+00:00"}