{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:L4K3CA7OY46STDIB2BTTG3BJP7","short_pith_number":"pith:L4K3CA7O","schema_version":"1.0","canonical_sha256":"5f15b103eec73d298d01d067336c297fcc98b95b422c9452bc7ba4494dad7df1","source":{"kind":"arxiv","id":"1803.02999","version":3},"attestation_state":"computed","paper":{"title":"On First-Order Meta-Learning Algorithms","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alex Nichol, John Schulman, Joshua Achiam","submitted_at":"2018-03-08T08:29:38Z","abstract_excerpt":"This paper considers meta-learning problems, where there is a distribution of tasks, and we would like to obtain an agent that performs well (i.e., learns quickly) when presented with a previously unseen task sampled from this distribution. We analyze a family of algorithms for learning a parameter initialization that can be fine-tuned quickly on a new task, using only first-order derivatives for the meta-learning updates. This family includes and generalizes first-order MAML, an approximation to MAML obtained by ignoring second-order derivatives. It also includes Reptile, a new algorithm that"},"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":"1803.02999","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-03-08T08:29:38Z","cross_cats_sorted":[],"title_canon_sha256":"c8f0406484169dc5e36717f8bd12d1a4a769a30bca2600590a6e3ec9a83a4560","abstract_canon_sha256":"e6913bbc1d8607eb146b2f143cb5f12f54736559963ecf63ead925a0a53d39cb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:02:45.135942Z","signature_b64":"5VR+LVW3VVtCpogIAikiMfrXrc+wbil4VKqrV21FGRTo2YXMw+ymysKcFwh/f85H372XehmwCbaH1xgK4XwCDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5f15b103eec73d298d01d067336c297fcc98b95b422c9452bc7ba4494dad7df1","last_reissued_at":"2026-05-18T00:02:45.135433Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:02:45.135433Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On First-Order Meta-Learning Algorithms","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alex Nichol, John Schulman, Joshua Achiam","submitted_at":"2018-03-08T08:29:38Z","abstract_excerpt":"This paper considers meta-learning problems, where there is a distribution of tasks, and we would like to obtain an agent that performs well (i.e., learns quickly) when presented with a previously unseen task sampled from this distribution. We analyze a family of algorithms for learning a parameter initialization that can be fine-tuned quickly on a new task, using only first-order derivatives for the meta-learning updates. This family includes and generalizes first-order MAML, an approximation to MAML obtained by ignoring second-order derivatives. It also includes Reptile, a new algorithm that"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1803.02999","kind":"arxiv","version":3},"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":"1803.02999","created_at":"2026-05-18T00:02:45.135493+00:00"},{"alias_kind":"arxiv_version","alias_value":"1803.02999v3","created_at":"2026-05-18T00:02:45.135493+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1803.02999","created_at":"2026-05-18T00:02:45.135493+00:00"},{"alias_kind":"pith_short_12","alias_value":"L4K3CA7OY46S","created_at":"2026-05-18T12:32:33.847187+00:00"},{"alias_kind":"pith_short_16","alias_value":"L4K3CA7OY46STDIB","created_at":"2026-05-18T12:32:33.847187+00:00"},{"alias_kind":"pith_short_8","alias_value":"L4K3CA7O","created_at":"2026-05-18T12:32:33.847187+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":29,"internal_anchor_count":21,"sample":[{"citing_arxiv_id":"1906.11415","citing_title":"Few-Shot Video Classification via Temporal Alignment","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22020","citing_title":"ForeSplat: Optimization-Aware Foresight for Feed-Forward 3D Gaussian Splatting","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2605.23595","citing_title":"Cost-Effective Model Evaluation with Meta-Learning","ref_index":65,"is_internal_anchor":true},{"citing_arxiv_id":"1907.05019","citing_title":"Massively Multilingual Neural Machine Translation in the Wild: Findings and Challenges","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"1907.05415","citing_title":"Learning to learn with quantum neural networks via classical neural networks","ref_index":61,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22020","citing_title":"ForeSplat: Optimization-Aware Foresight for Feed-Forward 3D Gaussian Splatting","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2502.07529","citing_title":"Training Deep Learning Models with Norm-Constrained LMOs","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20385","citing_title":"ConceptSeg-R1: Segment Any Concept via Meta-Reinforcement Learning","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20479","citing_title":"Oracle Supervision Transfers for Hyperparameter Prediction in Model-Based Image Denoising","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14133","citing_title":"ClawForge: Generating Executable Interactive Benchmarks for Command-Line Agents","ref_index":54,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16844","citing_title":"Artificial Adaptive Intelligence: The Missing Stage Between Narrow and General Intelligence","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2508.11205","citing_title":"Meta-learning Structure-Preserving Dynamics","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2509.22483","citing_title":"OFMU: Optimization-Driven Framework for Machine Unlearning","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2512.21204","citing_title":"SpidR-Adapt: A Universal Speech Representation Model for Few-Shot Adaptation","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2601.23179","citing_title":"Universal Adversarial Attacks against Closed-Source MLLMs via Target-View Routed Meta Optimization","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2602.07940","citing_title":"MePo: Meta Post-Refinement for Rehearsal-Free General Continual Learning","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2604.09576","citing_title":"AHC: Meta-Learned Adaptive Compression for Continual Object Detection on Memory-Constrained Microcontrollers","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14133","citing_title":"ClawForge: Generating Executable Interactive Benchmarks for Command-Line Agents","ref_index":54,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14527","citing_title":"Lang2MLIP: End-to-End Language-to-Machine Learning Interatomic Potential Development with Autonomous Agentic Workflows","ref_index":76,"is_internal_anchor":true},{"citing_arxiv_id":"2603.22241","citing_title":"MemDLM: Memory-Enhanced DLM Training","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2501.00663","citing_title":"Titans: Learning to Memorize at Test Time","ref_index":77,"is_internal_anchor":true},{"citing_arxiv_id":"2605.03279","citing_title":"RFPrompt: Prompt-Based Expert Adaptation of the Large Wireless Model for Modulation Classification","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26938","citing_title":"Meta-learning-enhanced implicit full waveform inversion","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09932","citing_title":"FocuSFT: Bilevel Optimization for Dilution-Aware Long-Context Fine-Tuning","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13130","citing_title":"Generalization Guarantees on Data-Driven Tuning of Gradient Descent with Langevin Updates","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L4K3CA7OY46STDIB2BTTG3BJP7","json":"https://pith.science/pith/L4K3CA7OY46STDIB2BTTG3BJP7.json","graph_json":"https://pith.science/api/pith-number/L4K3CA7OY46STDIB2BTTG3BJP7/graph.json","events_json":"https://pith.science/api/pith-number/L4K3CA7OY46STDIB2BTTG3BJP7/events.json","paper":"https://pith.science/paper/L4K3CA7O"},"agent_actions":{"view_html":"https://pith.science/pith/L4K3CA7OY46STDIB2BTTG3BJP7","download_json":"https://pith.science/pith/L4K3CA7OY46STDIB2BTTG3BJP7.json","view_paper":"https://pith.science/paper/L4K3CA7O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1803.02999&json=true","fetch_graph":"https://pith.science/api/pith-number/L4K3CA7OY46STDIB2BTTG3BJP7/graph.json","fetch_events":"https://pith.science/api/pith-number/L4K3CA7OY46STDIB2BTTG3BJP7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L4K3CA7OY46STDIB2BTTG3BJP7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L4K3CA7OY46STDIB2BTTG3BJP7/action/storage_attestation","attest_author":"https://pith.science/pith/L4K3CA7OY46STDIB2BTTG3BJP7/action/author_attestation","sign_citation":"https://pith.science/pith/L4K3CA7OY46STDIB2BTTG3BJP7/action/citation_signature","submit_replication":"https://pith.science/pith/L4K3CA7OY46STDIB2BTTG3BJP7/action/replication_record"}},"created_at":"2026-05-18T00:02:45.135493+00:00","updated_at":"2026-05-18T00:02:45.135493+00:00"}