{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:VD2WTM5IQDIHHJVBGMEPUH7UFL","short_pith_number":"pith:VD2WTM5I","schema_version":"1.0","canonical_sha256":"a8f569b3a880d073a6a13308fa1ff42af607e4732bfe84a657a81e1684426c62","source":{"kind":"arxiv","id":"1712.06563","version":3},"attestation_state":"computed","paper":{"title":"Safe Mutations for Deep and Recurrent Neural Networks through Output Gradients","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.NE","authors_text":"Jay Chen, Jeff Clune, Joel Lehman, Kenneth O. Stanley","submitted_at":"2017-12-18T18:16:51Z","abstract_excerpt":"While neuroevolution (evolving neural networks) has a successful track record across a variety of domains from reinforcement learning to artificial life, it is rarely applied to large, deep neural networks. A central reason is that while random mutation generally works in low dimensions, a random perturbation of thousands or millions of weights is likely to break existing functionality, providing no learning signal even if some individual weight changes were beneficial. This paper proposes a solution by introducing a family of safe mutation (SM) operators that aim within the mutation operator "},"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":"1712.06563","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2017-12-18T18:16:51Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7a0c020abb1f37a1dd83e31c2f94a57db79e520967cf40772db71475ed179c1b","abstract_canon_sha256":"5cf722a2d00aefe95bf91acc03d0ac7a134895a33e33f53ab1e81197806fab91"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:17:01.437771Z","signature_b64":"FtDfhVHSWAFdL0NNO9OkdezyysIFMdmJIpYGkeupYXZV5j++TzE5I6loxfYUiW7exMxJvIEVgufYLsPDGSn5Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a8f569b3a880d073a6a13308fa1ff42af607e4732bfe84a657a81e1684426c62","last_reissued_at":"2026-05-18T00:17:01.437246Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:17:01.437246Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Safe Mutations for Deep and Recurrent Neural Networks through Output Gradients","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.NE","authors_text":"Jay Chen, Jeff Clune, Joel Lehman, Kenneth O. Stanley","submitted_at":"2017-12-18T18:16:51Z","abstract_excerpt":"While neuroevolution (evolving neural networks) has a successful track record across a variety of domains from reinforcement learning to artificial life, it is rarely applied to large, deep neural networks. A central reason is that while random mutation generally works in low dimensions, a random perturbation of thousands or millions of weights is likely to break existing functionality, providing no learning signal even if some individual weight changes were beneficial. This paper proposes a solution by introducing a family of safe mutation (SM) operators that aim within the mutation operator "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1712.06563","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":"1712.06563","created_at":"2026-05-18T00:17:01.437330+00:00"},{"alias_kind":"arxiv_version","alias_value":"1712.06563v3","created_at":"2026-05-18T00:17:01.437330+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1712.06563","created_at":"2026-05-18T00:17:01.437330+00:00"},{"alias_kind":"pith_short_12","alias_value":"VD2WTM5IQDIH","created_at":"2026-05-18T12:31:49.984773+00:00"},{"alias_kind":"pith_short_16","alias_value":"VD2WTM5IQDIHHJVB","created_at":"2026-05-18T12:31:49.984773+00:00"},{"alias_kind":"pith_short_8","alias_value":"VD2WTM5I","created_at":"2026-05-18T12:31:49.984773+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VD2WTM5IQDIHHJVBGMEPUH7UFL","json":"https://pith.science/pith/VD2WTM5IQDIHHJVBGMEPUH7UFL.json","graph_json":"https://pith.science/api/pith-number/VD2WTM5IQDIHHJVBGMEPUH7UFL/graph.json","events_json":"https://pith.science/api/pith-number/VD2WTM5IQDIHHJVBGMEPUH7UFL/events.json","paper":"https://pith.science/paper/VD2WTM5I"},"agent_actions":{"view_html":"https://pith.science/pith/VD2WTM5IQDIHHJVBGMEPUH7UFL","download_json":"https://pith.science/pith/VD2WTM5IQDIHHJVBGMEPUH7UFL.json","view_paper":"https://pith.science/paper/VD2WTM5I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1712.06563&json=true","fetch_graph":"https://pith.science/api/pith-number/VD2WTM5IQDIHHJVBGMEPUH7UFL/graph.json","fetch_events":"https://pith.science/api/pith-number/VD2WTM5IQDIHHJVBGMEPUH7UFL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VD2WTM5IQDIHHJVBGMEPUH7UFL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VD2WTM5IQDIHHJVBGMEPUH7UFL/action/storage_attestation","attest_author":"https://pith.science/pith/VD2WTM5IQDIHHJVBGMEPUH7UFL/action/author_attestation","sign_citation":"https://pith.science/pith/VD2WTM5IQDIHHJVBGMEPUH7UFL/action/citation_signature","submit_replication":"https://pith.science/pith/VD2WTM5IQDIHHJVBGMEPUH7UFL/action/replication_record"}},"created_at":"2026-05-18T00:17:01.437330+00:00","updated_at":"2026-05-18T00:17:01.437330+00:00"}