{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:EFILPP7OKKILDBNJMWUV2UJF6U","short_pith_number":"pith:EFILPP7O","schema_version":"1.0","canonical_sha256":"2150b7bfee5290b185a965a95d5125f52bfd20240f7be2139d7c4b9d1b159878","source":{"kind":"arxiv","id":"1812.04604","version":1},"attestation_state":"computed","paper":{"title":"Diagnostic Visualization for Deep Neural Networks Using Stochastic Gradient Langevin Dynamics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Biye Jiang, David M. Chan, John F. Canny, Tianhao Zhang","submitted_at":"2018-12-11T18:43:52Z","abstract_excerpt":"The internal states of most deep neural networks are difficult to interpret, which makes diagnosis and debugging during training challenging. Activation maximization methods are widely used, but lead to multiple optima and are hard to interpret (appear noise-like) for complex neurons. Image-based methods use maximally-activating image regions which are easier to interpret, but do not provide pixel-level insight into why the neuron responds to them. In this work we introduce an MCMC method: Langevin Dynamics Activation Maximization (LDAM), which is designed for diagnostic visualization. LDAM pr"},"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":"1812.04604","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-12-11T18:43:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"adf5d3eaca2bdbf69affc2657784267fc8cbc39717d8fabc9a5dbbd21f49feef","abstract_canon_sha256":"0f83956bef6dd38fb84696c926a78393627075ef25bb29b894b502db43a0c03b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:58:31.607820Z","signature_b64":"+F5Fm9Etb1mL42D+FTu4TK7vF801pLZAf4alnnykNqN9oTGxcXIBCCtlhtd2zMjuJ0IovD1gLs3gVdoX7YjcBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2150b7bfee5290b185a965a95d5125f52bfd20240f7be2139d7c4b9d1b159878","last_reissued_at":"2026-05-17T23:58:31.607090Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:58:31.607090Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diagnostic Visualization for Deep Neural Networks Using Stochastic Gradient Langevin Dynamics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Biye Jiang, David M. Chan, John F. Canny, Tianhao Zhang","submitted_at":"2018-12-11T18:43:52Z","abstract_excerpt":"The internal states of most deep neural networks are difficult to interpret, which makes diagnosis and debugging during training challenging. Activation maximization methods are widely used, but lead to multiple optima and are hard to interpret (appear noise-like) for complex neurons. Image-based methods use maximally-activating image regions which are easier to interpret, but do not provide pixel-level insight into why the neuron responds to them. In this work we introduce an MCMC method: Langevin Dynamics Activation Maximization (LDAM), which is designed for diagnostic visualization. LDAM pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1812.04604","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":"1812.04604","created_at":"2026-05-17T23:58:31.607218+00:00"},{"alias_kind":"arxiv_version","alias_value":"1812.04604v1","created_at":"2026-05-17T23:58:31.607218+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1812.04604","created_at":"2026-05-17T23:58:31.607218+00:00"},{"alias_kind":"pith_short_12","alias_value":"EFILPP7OKKIL","created_at":"2026-05-18T12:32:22.470017+00:00"},{"alias_kind":"pith_short_16","alias_value":"EFILPP7OKKILDBNJ","created_at":"2026-05-18T12:32:22.470017+00:00"},{"alias_kind":"pith_short_8","alias_value":"EFILPP7O","created_at":"2026-05-18T12:32:22.470017+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/EFILPP7OKKILDBNJMWUV2UJF6U","json":"https://pith.science/pith/EFILPP7OKKILDBNJMWUV2UJF6U.json","graph_json":"https://pith.science/api/pith-number/EFILPP7OKKILDBNJMWUV2UJF6U/graph.json","events_json":"https://pith.science/api/pith-number/EFILPP7OKKILDBNJMWUV2UJF6U/events.json","paper":"https://pith.science/paper/EFILPP7O"},"agent_actions":{"view_html":"https://pith.science/pith/EFILPP7OKKILDBNJMWUV2UJF6U","download_json":"https://pith.science/pith/EFILPP7OKKILDBNJMWUV2UJF6U.json","view_paper":"https://pith.science/paper/EFILPP7O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1812.04604&json=true","fetch_graph":"https://pith.science/api/pith-number/EFILPP7OKKILDBNJMWUV2UJF6U/graph.json","fetch_events":"https://pith.science/api/pith-number/EFILPP7OKKILDBNJMWUV2UJF6U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EFILPP7OKKILDBNJMWUV2UJF6U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EFILPP7OKKILDBNJMWUV2UJF6U/action/storage_attestation","attest_author":"https://pith.science/pith/EFILPP7OKKILDBNJMWUV2UJF6U/action/author_attestation","sign_citation":"https://pith.science/pith/EFILPP7OKKILDBNJMWUV2UJF6U/action/citation_signature","submit_replication":"https://pith.science/pith/EFILPP7OKKILDBNJMWUV2UJF6U/action/replication_record"}},"created_at":"2026-05-17T23:58:31.607218+00:00","updated_at":"2026-05-17T23:58:31.607218+00:00"}