{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2013:FT3W22ZBGHMTSYTRIUNPIJSU7U","short_pith_number":"pith:FT3W22ZB","schema_version":"1.0","canonical_sha256":"2cf76d6b2131d9396271451af42654fd0e69c2f87c0976b7bb13b0449a2b51af","source":{"kind":"arxiv","id":"1305.2982","version":1},"attestation_state":"computed","paper":{"title":"Estimating or Propagating Gradients Through Stochastic Neurons","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Yoshua Bengio","submitted_at":"2013-05-14T00:29:42Z","abstract_excerpt":"Stochastic neurons can be useful for a number of reasons in deep learning models, but in many cases they pose a challenging problem: how to estimate the gradient of a loss function with respect to the input of such stochastic neurons, i.e., can we \"back-propagate\" through these stochastic neurons? We examine this question, existing approaches, and present two novel families of solutions, applicable in different settings. In particular, it is demonstrated that a simple biologically plausible formula gives rise to an an unbiased (but noisy) estimator of the gradient with respect to a binary stoc"},"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":"1305.2982","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2013-05-14T00:29:42Z","cross_cats_sorted":[],"title_canon_sha256":"f7c4ee353b6369d8bdb79100fdf4da9a6a0f5fb55870e1b240a732d995530256","abstract_canon_sha256":"ea96bf37ca5f9e51c1a0ceab865c4b856703ab3aa02270497aad3a1244df28b0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:25:45.258690Z","signature_b64":"x239iQJZS/J+rO5TWptXTjfMBwDcvBmAAEk/iuKbc3Gd4nYc/TrP+v1K33SdW8H7ukY2bxFsdwjtEqxiH+KIBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2cf76d6b2131d9396271451af42654fd0e69c2f87c0976b7bb13b0449a2b51af","last_reissued_at":"2026-05-18T03:25:45.257770Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:25:45.257770Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Estimating or Propagating Gradients Through Stochastic Neurons","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Yoshua Bengio","submitted_at":"2013-05-14T00:29:42Z","abstract_excerpt":"Stochastic neurons can be useful for a number of reasons in deep learning models, but in many cases they pose a challenging problem: how to estimate the gradient of a loss function with respect to the input of such stochastic neurons, i.e., can we \"back-propagate\" through these stochastic neurons? We examine this question, existing approaches, and present two novel families of solutions, applicable in different settings. In particular, it is demonstrated that a simple biologically plausible formula gives rise to an an unbiased (but noisy) estimator of the gradient with respect to a binary stoc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1305.2982","kind":"arxiv","version":1},"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":"1305.2982","created_at":"2026-05-18T03:25:45.257929+00:00"},{"alias_kind":"arxiv_version","alias_value":"1305.2982v1","created_at":"2026-05-18T03:25:45.257929+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1305.2982","created_at":"2026-05-18T03:25:45.257929+00:00"},{"alias_kind":"pith_short_12","alias_value":"FT3W22ZBGHMT","created_at":"2026-05-18T12:27:45.050594+00:00"},{"alias_kind":"pith_short_16","alias_value":"FT3W22ZBGHMTSYTR","created_at":"2026-05-18T12:27:45.050594+00:00"},{"alias_kind":"pith_short_8","alias_value":"FT3W22ZB","created_at":"2026-05-18T12:27:45.050594+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"1906.09992","citing_title":"Learning Latent Trees with Stochastic Perturbations and Differentiable Dynamic Programming","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18624","citing_title":"Learning to Look Benign: Targeted Evasion of Malware Detectors via API Import Injection","ref_index":37,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FT3W22ZBGHMTSYTRIUNPIJSU7U","json":"https://pith.science/pith/FT3W22ZBGHMTSYTRIUNPIJSU7U.json","graph_json":"https://pith.science/api/pith-number/FT3W22ZBGHMTSYTRIUNPIJSU7U/graph.json","events_json":"https://pith.science/api/pith-number/FT3W22ZBGHMTSYTRIUNPIJSU7U/events.json","paper":"https://pith.science/paper/FT3W22ZB"},"agent_actions":{"view_html":"https://pith.science/pith/FT3W22ZBGHMTSYTRIUNPIJSU7U","download_json":"https://pith.science/pith/FT3W22ZBGHMTSYTRIUNPIJSU7U.json","view_paper":"https://pith.science/paper/FT3W22ZB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1305.2982&json=true","fetch_graph":"https://pith.science/api/pith-number/FT3W22ZBGHMTSYTRIUNPIJSU7U/graph.json","fetch_events":"https://pith.science/api/pith-number/FT3W22ZBGHMTSYTRIUNPIJSU7U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FT3W22ZBGHMTSYTRIUNPIJSU7U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FT3W22ZBGHMTSYTRIUNPIJSU7U/action/storage_attestation","attest_author":"https://pith.science/pith/FT3W22ZBGHMTSYTRIUNPIJSU7U/action/author_attestation","sign_citation":"https://pith.science/pith/FT3W22ZBGHMTSYTRIUNPIJSU7U/action/citation_signature","submit_replication":"https://pith.science/pith/FT3W22ZBGHMTSYTRIUNPIJSU7U/action/replication_record"}},"created_at":"2026-05-18T03:25:45.257929+00:00","updated_at":"2026-05-18T03:25:45.257929+00:00"}