{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:5FDLBD5DVBEJPOUPMHMVE4HCWK","short_pith_number":"pith:5FDLBD5D","schema_version":"1.0","canonical_sha256":"e946b08fa3a84897ba8f61d95270e2b2931f6701fa5c732a92f5e9abddc73a17","source":{"kind":"arxiv","id":"2106.02749","version":2},"attestation_state":"computed","paper":{"title":"Predify: Augmenting deep neural networks with brain-inspired predictive coding dynamics","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["q-bio.NC"],"primary_cat":"cs.CV","authors_text":"Andrea Alamia, Benjamin Ador, Bhavin Choksi, Callum Biggs O'May, Milad Mozafari, Rufin VanRullen","submitted_at":"2021-06-04T22:48:13Z","abstract_excerpt":"Deep neural networks excel at image classification, but their performance is far less robust to input perturbations than human perception. In this work we explore whether this shortcoming may be partly addressed by incorporating brain-inspired recurrent dynamics in deep convolutional networks. We take inspiration from a popular framework in neuroscience: 'predictive coding'. At each layer of the hierarchical model, generative feedback 'predicts' (i.e., reconstructs) the pattern of activity in the previous layer. The reconstruction errors are used to iteratively update the network's representat"},"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":"2106.02749","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-06-04T22:48:13Z","cross_cats_sorted":["q-bio.NC"],"title_canon_sha256":"580c936ca78dd254cc0cfe8b41250d8773d4317a67bb441ceb216532c6e1e104","abstract_canon_sha256":"0e63d9fc8b8988b35f5713256f87deda8e47ec4db30e81156ee70280b32f908d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:29:01.435041Z","signature_b64":"/DAA+QfQPxB1PnrDFWhw84ZZ7UNb6av6A9I7wyC2jCT+2zQiu+xczbHJebqwEU5FeRLhC9QMd1Z5To7f8ZgkAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e946b08fa3a84897ba8f61d95270e2b2931f6701fa5c732a92f5e9abddc73a17","last_reissued_at":"2026-07-05T03:29:01.434622Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:29:01.434622Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Predify: Augmenting deep neural networks with brain-inspired predictive coding dynamics","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["q-bio.NC"],"primary_cat":"cs.CV","authors_text":"Andrea Alamia, Benjamin Ador, Bhavin Choksi, Callum Biggs O'May, Milad Mozafari, Rufin VanRullen","submitted_at":"2021-06-04T22:48:13Z","abstract_excerpt":"Deep neural networks excel at image classification, but their performance is far less robust to input perturbations than human perception. In this work we explore whether this shortcoming may be partly addressed by incorporating brain-inspired recurrent dynamics in deep convolutional networks. We take inspiration from a popular framework in neuroscience: 'predictive coding'. At each layer of the hierarchical model, generative feedback 'predicts' (i.e., reconstructs) the pattern of activity in the previous layer. The reconstruction errors are used to iteratively update the network's representat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.02749","kind":"arxiv","version":2},"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/2106.02749/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":"2106.02749","created_at":"2026-07-05T03:29:01.434680+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.02749v2","created_at":"2026-07-05T03:29:01.434680+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.02749","created_at":"2026-07-05T03:29:01.434680+00:00"},{"alias_kind":"pith_short_12","alias_value":"5FDLBD5DVBEJ","created_at":"2026-07-05T03:29:01.434680+00:00"},{"alias_kind":"pith_short_16","alias_value":"5FDLBD5DVBEJPOUP","created_at":"2026-07-05T03:29:01.434680+00:00"},{"alias_kind":"pith_short_8","alias_value":"5FDLBD5D","created_at":"2026-07-05T03:29:01.434680+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.17737","citing_title":"Contextual Feedback Loops: Amplifying Deep Reasoning with Iterative Top-Down Feedback","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5FDLBD5DVBEJPOUPMHMVE4HCWK","json":"https://pith.science/pith/5FDLBD5DVBEJPOUPMHMVE4HCWK.json","graph_json":"https://pith.science/api/pith-number/5FDLBD5DVBEJPOUPMHMVE4HCWK/graph.json","events_json":"https://pith.science/api/pith-number/5FDLBD5DVBEJPOUPMHMVE4HCWK/events.json","paper":"https://pith.science/paper/5FDLBD5D"},"agent_actions":{"view_html":"https://pith.science/pith/5FDLBD5DVBEJPOUPMHMVE4HCWK","download_json":"https://pith.science/pith/5FDLBD5DVBEJPOUPMHMVE4HCWK.json","view_paper":"https://pith.science/paper/5FDLBD5D","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.02749&json=true","fetch_graph":"https://pith.science/api/pith-number/5FDLBD5DVBEJPOUPMHMVE4HCWK/graph.json","fetch_events":"https://pith.science/api/pith-number/5FDLBD5DVBEJPOUPMHMVE4HCWK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5FDLBD5DVBEJPOUPMHMVE4HCWK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5FDLBD5DVBEJPOUPMHMVE4HCWK/action/storage_attestation","attest_author":"https://pith.science/pith/5FDLBD5DVBEJPOUPMHMVE4HCWK/action/author_attestation","sign_citation":"https://pith.science/pith/5FDLBD5DVBEJPOUPMHMVE4HCWK/action/citation_signature","submit_replication":"https://pith.science/pith/5FDLBD5DVBEJPOUPMHMVE4HCWK/action/replication_record"}},"created_at":"2026-07-05T03:29:01.434680+00:00","updated_at":"2026-07-05T03:29:01.434680+00:00"}