{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:WY7TBCZ777LGHWRAJXWKBEFL44","short_pith_number":"pith:WY7TBCZ7","schema_version":"1.0","canonical_sha256":"b63f308b3fffd663da204deca090abe728ac9e34ddd49df1efc72eaed21d1262","source":{"kind":"arxiv","id":"2112.09214","version":2},"attestation_state":"computed","paper":{"title":"Sparse Coding with Multi-Layer Decoders using Variance Regularization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Katrina Evtimova, Yann LeCun","submitted_at":"2021-12-16T21:46:23Z","abstract_excerpt":"Sparse representations of images are useful in many computer vision applications. Sparse coding with an $l_1$ penalty and a learned linear dictionary requires regularization of the dictionary to prevent a collapse in the $l_1$ norms of the codes. Typically, this regularization entails bounding the Euclidean norms of the dictionary's elements. In this work, we propose a novel sparse coding protocol which prevents a collapse in the codes without the need to regularize the decoder. Our method regularizes the codes directly so that each latent code component has variance greater than a fixed thres"},"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":"2112.09214","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-12-16T21:46:23Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2d53eeb09e51d26761cd2a88faeb1367929d96658d5a253dfe35aa3301007d1e","abstract_canon_sha256":"476ec52df97b9890be15892caf9a8e7448f5ee7c760e93c2211f8431ea892237"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:55:37.380427Z","signature_b64":"GO92Dyy0pOW2KCSCEBNaT8XDsmRVbRIUj6sA3n/WMnGyYV5d5yAOZclhCHtCBA7Lj2zCKWg9G1+ykr0Jo4kDBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b63f308b3fffd663da204deca090abe728ac9e34ddd49df1efc72eaed21d1262","last_reissued_at":"2026-07-05T04:55:37.379966Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:55:37.379966Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sparse Coding with Multi-Layer Decoders using Variance Regularization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Katrina Evtimova, Yann LeCun","submitted_at":"2021-12-16T21:46:23Z","abstract_excerpt":"Sparse representations of images are useful in many computer vision applications. Sparse coding with an $l_1$ penalty and a learned linear dictionary requires regularization of the dictionary to prevent a collapse in the $l_1$ norms of the codes. Typically, this regularization entails bounding the Euclidean norms of the dictionary's elements. In this work, we propose a novel sparse coding protocol which prevents a collapse in the codes without the need to regularize the decoder. Our method regularizes the codes directly so that each latent code component has variance greater than a fixed thres"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.09214","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/2112.09214/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":"2112.09214","created_at":"2026-07-05T04:55:37.380022+00:00"},{"alias_kind":"arxiv_version","alias_value":"2112.09214v2","created_at":"2026-07-05T04:55:37.380022+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.09214","created_at":"2026-07-05T04:55:37.380022+00:00"},{"alias_kind":"pith_short_12","alias_value":"WY7TBCZ777LG","created_at":"2026-07-05T04:55:37.380022+00:00"},{"alias_kind":"pith_short_16","alias_value":"WY7TBCZ777LGHWRA","created_at":"2026-07-05T04:55:37.380022+00:00"},{"alias_kind":"pith_short_8","alias_value":"WY7TBCZ7","created_at":"2026-07-05T04:55:37.380022+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/WY7TBCZ777LGHWRAJXWKBEFL44","json":"https://pith.science/pith/WY7TBCZ777LGHWRAJXWKBEFL44.json","graph_json":"https://pith.science/api/pith-number/WY7TBCZ777LGHWRAJXWKBEFL44/graph.json","events_json":"https://pith.science/api/pith-number/WY7TBCZ777LGHWRAJXWKBEFL44/events.json","paper":"https://pith.science/paper/WY7TBCZ7"},"agent_actions":{"view_html":"https://pith.science/pith/WY7TBCZ777LGHWRAJXWKBEFL44","download_json":"https://pith.science/pith/WY7TBCZ777LGHWRAJXWKBEFL44.json","view_paper":"https://pith.science/paper/WY7TBCZ7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2112.09214&json=true","fetch_graph":"https://pith.science/api/pith-number/WY7TBCZ777LGHWRAJXWKBEFL44/graph.json","fetch_events":"https://pith.science/api/pith-number/WY7TBCZ777LGHWRAJXWKBEFL44/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WY7TBCZ777LGHWRAJXWKBEFL44/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WY7TBCZ777LGHWRAJXWKBEFL44/action/storage_attestation","attest_author":"https://pith.science/pith/WY7TBCZ777LGHWRAJXWKBEFL44/action/author_attestation","sign_citation":"https://pith.science/pith/WY7TBCZ777LGHWRAJXWKBEFL44/action/citation_signature","submit_replication":"https://pith.science/pith/WY7TBCZ777LGHWRAJXWKBEFL44/action/replication_record"}},"created_at":"2026-07-05T04:55:37.380022+00:00","updated_at":"2026-07-05T04:55:37.380022+00:00"}