{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:JDJEP6ROYPU5ZOXPIJYLZHS4VR","short_pith_number":"pith:JDJEP6RO","schema_version":"1.0","canonical_sha256":"48d247fa2ec3e9dcbaef4270bc9e5cac51af31ae9778e786812615f5557bda85","source":{"kind":"arxiv","id":"2002.09815","version":3},"attestation_state":"computed","paper":{"title":"Neuron Shapley: Discovering the Responsible Neurons","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.LG","cs.NE"],"primary_cat":"stat.ML","authors_text":"Amirata Ghorbani, James Zou","submitted_at":"2020-02-23T03:29:58Z","abstract_excerpt":"We develop Neuron Shapley as a new framework to quantify the contribution of individual neurons to the prediction and performance of a deep network. By accounting for interactions across neurons, Neuron Shapley is more effective in identifying important filters compared to common approaches based on activation patterns. Interestingly, removing just 30 filters with the highest Shapley scores effectively destroys the prediction accuracy of Inception-v3 on ImageNet. Visualization of these few critical filters provides insights into how the network functions. Neuron Shapley is a flexible framework"},"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":"2002.09815","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-02-23T03:29:58Z","cross_cats_sorted":["cs.CV","cs.LG","cs.NE"],"title_canon_sha256":"88e50027602d3f58cbfcf848faacc5f5cc3e54450682ea01bf1f44e7c290734f","abstract_canon_sha256":"cf24fe0d6b7448cf329e00f2e05a0a4df699cba695be5d917abb918c056b15b1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:51:27.740681Z","signature_b64":"poShsbMADARn3O1AGcSOpmsJSyEOsrPro4BekI+7DHRwXA2OWzSlU4N5f6wPQGbRm1WAoTs6OwbTWluqXjGrDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"48d247fa2ec3e9dcbaef4270bc9e5cac51af31ae9778e786812615f5557bda85","last_reissued_at":"2026-07-05T01:51:27.740277Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:51:27.740277Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neuron Shapley: Discovering the Responsible Neurons","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.LG","cs.NE"],"primary_cat":"stat.ML","authors_text":"Amirata Ghorbani, James Zou","submitted_at":"2020-02-23T03:29:58Z","abstract_excerpt":"We develop Neuron Shapley as a new framework to quantify the contribution of individual neurons to the prediction and performance of a deep network. By accounting for interactions across neurons, Neuron Shapley is more effective in identifying important filters compared to common approaches based on activation patterns. Interestingly, removing just 30 filters with the highest Shapley scores effectively destroys the prediction accuracy of Inception-v3 on ImageNet. Visualization of these few critical filters provides insights into how the network functions. Neuron Shapley is a flexible framework"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.09815","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2002.09815/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":"2002.09815","created_at":"2026-07-05T01:51:27.740340+00:00"},{"alias_kind":"arxiv_version","alias_value":"2002.09815v3","created_at":"2026-07-05T01:51:27.740340+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.09815","created_at":"2026-07-05T01:51:27.740340+00:00"},{"alias_kind":"pith_short_12","alias_value":"JDJEP6ROYPU5","created_at":"2026-07-05T01:51:27.740340+00:00"},{"alias_kind":"pith_short_16","alias_value":"JDJEP6ROYPU5ZOXP","created_at":"2026-07-05T01:51:27.740340+00:00"},{"alias_kind":"pith_short_8","alias_value":"JDJEP6RO","created_at":"2026-07-05T01:51:27.740340+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.16109","citing_title":"From Indirect Object Identification to Syllogisms: Exploring Binary Mechanisms in Transformer Circuits","ref_index":18,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JDJEP6ROYPU5ZOXPIJYLZHS4VR","json":"https://pith.science/pith/JDJEP6ROYPU5ZOXPIJYLZHS4VR.json","graph_json":"https://pith.science/api/pith-number/JDJEP6ROYPU5ZOXPIJYLZHS4VR/graph.json","events_json":"https://pith.science/api/pith-number/JDJEP6ROYPU5ZOXPIJYLZHS4VR/events.json","paper":"https://pith.science/paper/JDJEP6RO"},"agent_actions":{"view_html":"https://pith.science/pith/JDJEP6ROYPU5ZOXPIJYLZHS4VR","download_json":"https://pith.science/pith/JDJEP6ROYPU5ZOXPIJYLZHS4VR.json","view_paper":"https://pith.science/paper/JDJEP6RO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2002.09815&json=true","fetch_graph":"https://pith.science/api/pith-number/JDJEP6ROYPU5ZOXPIJYLZHS4VR/graph.json","fetch_events":"https://pith.science/api/pith-number/JDJEP6ROYPU5ZOXPIJYLZHS4VR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JDJEP6ROYPU5ZOXPIJYLZHS4VR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JDJEP6ROYPU5ZOXPIJYLZHS4VR/action/storage_attestation","attest_author":"https://pith.science/pith/JDJEP6ROYPU5ZOXPIJYLZHS4VR/action/author_attestation","sign_citation":"https://pith.science/pith/JDJEP6ROYPU5ZOXPIJYLZHS4VR/action/citation_signature","submit_replication":"https://pith.science/pith/JDJEP6ROYPU5ZOXPIJYLZHS4VR/action/replication_record"}},"created_at":"2026-07-05T01:51:27.740340+00:00","updated_at":"2026-07-05T01:51:27.740340+00:00"}