{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AFFW6FX7QHYSZJ3TMAPVNLQX7Z","short_pith_number":"pith:AFFW6FX7","schema_version":"1.0","canonical_sha256":"014b6f16ff81f12ca773601f56ae17fe582927fc9fa35c47abff6a0fec65841c","source":{"kind":"arxiv","id":"2410.20634","version":1},"attestation_state":"computed","paper":{"title":"Plastic Learning with Deep Fourier Features","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alex Lewandowski, Dale Schuurmans, Marlos C. Machado","submitted_at":"2024-10-27T23:38:06Z","abstract_excerpt":"Deep neural networks can struggle to learn continually in the face of non-stationarity. This phenomenon is known as loss of plasticity. In this paper, we identify underlying principles that lead to plastic algorithms. In particular, we provide theoretical results showing that linear function approximation, as well as a special case of deep linear networks, do not suffer from loss of plasticity. We then propose deep Fourier features, which are the concatenation of a sine and cosine in every layer, and we show that this combination provides a dynamic balance between the trainability obtained thr"},"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":"2410.20634","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-27T23:38:06Z","cross_cats_sorted":[],"title_canon_sha256":"a361449acf2d486e57d5dd009ae0d0edd2186d4689a23fa6b90e64d6bfb1e3d7","abstract_canon_sha256":"2ffcb0c1b2702bc13530bb0d271fd59e4a34027461a918d736c84551d8bc8b02"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:27:08.503877Z","signature_b64":"KJLmLYKZcmXc01xo+3xu6xwQbnOoEPORjlT0KOs7q4Ds/ybqvVOty+f4zfUJnbMGSVouvWtiyX1wTcYlbcteAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"014b6f16ff81f12ca773601f56ae17fe582927fc9fa35c47abff6a0fec65841c","last_reissued_at":"2026-07-05T09:27:08.503421Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:27:08.503421Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Plastic Learning with Deep Fourier Features","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alex Lewandowski, Dale Schuurmans, Marlos C. Machado","submitted_at":"2024-10-27T23:38:06Z","abstract_excerpt":"Deep neural networks can struggle to learn continually in the face of non-stationarity. This phenomenon is known as loss of plasticity. In this paper, we identify underlying principles that lead to plastic algorithms. In particular, we provide theoretical results showing that linear function approximation, as well as a special case of deep linear networks, do not suffer from loss of plasticity. We then propose deep Fourier features, which are the concatenation of a sine and cosine in every layer, and we show that this combination provides a dynamic balance between the trainability obtained thr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.20634","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2410.20634/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":"2410.20634","created_at":"2026-07-05T09:27:08.503470+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.20634v1","created_at":"2026-07-05T09:27:08.503470+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.20634","created_at":"2026-07-05T09:27:08.503470+00:00"},{"alias_kind":"pith_short_12","alias_value":"AFFW6FX7QHYS","created_at":"2026-07-05T09:27:08.503470+00:00"},{"alias_kind":"pith_short_16","alias_value":"AFFW6FX7QHYSZJ3T","created_at":"2026-07-05T09:27:08.503470+00:00"},{"alias_kind":"pith_short_8","alias_value":"AFFW6FX7","created_at":"2026-07-05T09:27:08.503470+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.01342","citing_title":"Activation by Interval-wise Dropout: A Simple Way to Prevent Neural Networks from Plasticity Loss","ref_index":30,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AFFW6FX7QHYSZJ3TMAPVNLQX7Z","json":"https://pith.science/pith/AFFW6FX7QHYSZJ3TMAPVNLQX7Z.json","graph_json":"https://pith.science/api/pith-number/AFFW6FX7QHYSZJ3TMAPVNLQX7Z/graph.json","events_json":"https://pith.science/api/pith-number/AFFW6FX7QHYSZJ3TMAPVNLQX7Z/events.json","paper":"https://pith.science/paper/AFFW6FX7"},"agent_actions":{"view_html":"https://pith.science/pith/AFFW6FX7QHYSZJ3TMAPVNLQX7Z","download_json":"https://pith.science/pith/AFFW6FX7QHYSZJ3TMAPVNLQX7Z.json","view_paper":"https://pith.science/paper/AFFW6FX7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.20634&json=true","fetch_graph":"https://pith.science/api/pith-number/AFFW6FX7QHYSZJ3TMAPVNLQX7Z/graph.json","fetch_events":"https://pith.science/api/pith-number/AFFW6FX7QHYSZJ3TMAPVNLQX7Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AFFW6FX7QHYSZJ3TMAPVNLQX7Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AFFW6FX7QHYSZJ3TMAPVNLQX7Z/action/storage_attestation","attest_author":"https://pith.science/pith/AFFW6FX7QHYSZJ3TMAPVNLQX7Z/action/author_attestation","sign_citation":"https://pith.science/pith/AFFW6FX7QHYSZJ3TMAPVNLQX7Z/action/citation_signature","submit_replication":"https://pith.science/pith/AFFW6FX7QHYSZJ3TMAPVNLQX7Z/action/replication_record"}},"created_at":"2026-07-05T09:27:08.503470+00:00","updated_at":"2026-07-05T09:27:08.503470+00:00"}