{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:POMIHO76WGTYI2YD7CZ7I4VJOY","short_pith_number":"pith:POMIHO76","schema_version":"1.0","canonical_sha256":"7b9883bbfeb1a7846b03f8b3f472a9761bf1b2b6a2cfa3632fbd2f51db9f18e2","source":{"kind":"arxiv","id":"1903.11257","version":2},"attestation_state":"computed","paper":{"title":"How Can We Be So Dense? The Benefits of Using Highly Sparse Representations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Luiz Scheinkman, Subutai Ahmad","submitted_at":"2019-03-27T05:43:33Z","abstract_excerpt":"Most artificial networks today rely on dense representations, whereas biological networks rely on sparse representations. In this paper we show how sparse representations can be more robust to noise and interference, as long as the underlying dimensionality is sufficiently high. A key intuition that we develop is that the ratio of the operable volume around a sparse vector divided by the volume of the representational space decreases exponentially with dimensionality. We then analyze computationally efficient sparse networks containing both sparse weights and activations. Simulations on MNIST "},"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":"1903.11257","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2019-03-27T05:43:33Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"05de051411dbfae843cb1f7a4fe1c84e88e7b113145232801b020e82285e4f76","abstract_canon_sha256":"723f47ef217582450e79c199a103b2e80865d9fa139a6ae80ba976245498d464"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:49:38.343939Z","signature_b64":"Vmghc+SVutPTQtaK4NcD2hQZSV2+G01t2Cw3xYi4iRKjKFp/iQSTLeMJ9Am2VcFXFzYEjg7BUwo6hzJnOHdXCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7b9883bbfeb1a7846b03f8b3f472a9761bf1b2b6a2cfa3632fbd2f51db9f18e2","last_reissued_at":"2026-05-17T23:49:38.343261Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:49:38.343261Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Can We Be So Dense? The Benefits of Using Highly Sparse Representations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Luiz Scheinkman, Subutai Ahmad","submitted_at":"2019-03-27T05:43:33Z","abstract_excerpt":"Most artificial networks today rely on dense representations, whereas biological networks rely on sparse representations. In this paper we show how sparse representations can be more robust to noise and interference, as long as the underlying dimensionality is sufficiently high. A key intuition that we develop is that the ratio of the operable volume around a sparse vector divided by the volume of the representational space decreases exponentially with dimensionality. We then analyze computationally efficient sparse networks containing both sparse weights and activations. Simulations on MNIST "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.11257","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":""},"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":"1903.11257","created_at":"2026-05-17T23:49:38.343383+00:00"},{"alias_kind":"arxiv_version","alias_value":"1903.11257v2","created_at":"2026-05-17T23:49:38.343383+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.11257","created_at":"2026-05-17T23:49:38.343383+00:00"},{"alias_kind":"pith_short_12","alias_value":"POMIHO76WGTY","created_at":"2026-05-18T12:33:24.271573+00:00"},{"alias_kind":"pith_short_16","alias_value":"POMIHO76WGTYI2YD","created_at":"2026-05-18T12:33:24.271573+00:00"},{"alias_kind":"pith_short_8","alias_value":"POMIHO76","created_at":"2026-05-18T12:33:24.271573+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.13081","citing_title":"Selectivity and Shape in the Design of Forward-Forward Goodness Functions","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/POMIHO76WGTYI2YD7CZ7I4VJOY","json":"https://pith.science/pith/POMIHO76WGTYI2YD7CZ7I4VJOY.json","graph_json":"https://pith.science/api/pith-number/POMIHO76WGTYI2YD7CZ7I4VJOY/graph.json","events_json":"https://pith.science/api/pith-number/POMIHO76WGTYI2YD7CZ7I4VJOY/events.json","paper":"https://pith.science/paper/POMIHO76"},"agent_actions":{"view_html":"https://pith.science/pith/POMIHO76WGTYI2YD7CZ7I4VJOY","download_json":"https://pith.science/pith/POMIHO76WGTYI2YD7CZ7I4VJOY.json","view_paper":"https://pith.science/paper/POMIHO76","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1903.11257&json=true","fetch_graph":"https://pith.science/api/pith-number/POMIHO76WGTYI2YD7CZ7I4VJOY/graph.json","fetch_events":"https://pith.science/api/pith-number/POMIHO76WGTYI2YD7CZ7I4VJOY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/POMIHO76WGTYI2YD7CZ7I4VJOY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/POMIHO76WGTYI2YD7CZ7I4VJOY/action/storage_attestation","attest_author":"https://pith.science/pith/POMIHO76WGTYI2YD7CZ7I4VJOY/action/author_attestation","sign_citation":"https://pith.science/pith/POMIHO76WGTYI2YD7CZ7I4VJOY/action/citation_signature","submit_replication":"https://pith.science/pith/POMIHO76WGTYI2YD7CZ7I4VJOY/action/replication_record"}},"created_at":"2026-05-17T23:49:38.343383+00:00","updated_at":"2026-05-17T23:49:38.343383+00:00"}