{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2015:7OSXHIMA5Q4H5XAHE4QMHRANHM","short_pith_number":"pith:7OSXHIMA","schema_version":"1.0","canonical_sha256":"fba573a180ec387edc072720c3c40d3b10679da9a1fe1298f466b94236280f6e","source":{"kind":"arxiv","id":"1507.00210","version":1},"attestation_state":"computed","paper":{"title":"Natural Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.NE"],"primary_cat":"stat.ML","authors_text":"Guillaume Desjardins, Karen Simonyan, Koray Kavukcuoglu, Razvan Pascanu","submitted_at":"2015-07-01T12:42:01Z","abstract_excerpt":"We introduce Natural Neural Networks, a novel family of algorithms that speed up convergence by adapting their internal representation during training to improve conditioning of the Fisher matrix. In particular, we show a specific example that employs a simple and efficient reparametrization of the neural network weights by implicitly whitening the representation obtained at each layer, while preserving the feed-forward computation of the network. Such networks can be trained efficiently via the proposed Projected Natural Gradient Descent algorithm (PRONG), which amortizes the cost of these re"},"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":"1507.00210","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2015-07-01T12:42:01Z","cross_cats_sorted":["cs.LG","cs.NE"],"title_canon_sha256":"b342aa5d032e5aaf920f7fb435eef0c97e9fbe8e7f4d151ecabd4f0f22233428","abstract_canon_sha256":"786b9d39356a19452cf3668f682fa080206b37f4e23c50e44520eb13df2641a9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:37:29.561937Z","signature_b64":"KuUFYoKSZJSeslxu1juaKofOS8GT3DkoQ+iWEEGwaBv3XiFaIzqd0KxMqmHTK0daZYXeZ/zDK6RVVcTvqQ3/Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fba573a180ec387edc072720c3c40d3b10679da9a1fe1298f466b94236280f6e","last_reissued_at":"2026-05-18T01:37:29.561333Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:37:29.561333Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Natural Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.NE"],"primary_cat":"stat.ML","authors_text":"Guillaume Desjardins, Karen Simonyan, Koray Kavukcuoglu, Razvan Pascanu","submitted_at":"2015-07-01T12:42:01Z","abstract_excerpt":"We introduce Natural Neural Networks, a novel family of algorithms that speed up convergence by adapting their internal representation during training to improve conditioning of the Fisher matrix. In particular, we show a specific example that employs a simple and efficient reparametrization of the neural network weights by implicitly whitening the representation obtained at each layer, while preserving the feed-forward computation of the network. Such networks can be trained efficiently via the proposed Projected Natural Gradient Descent algorithm (PRONG), which amortizes the cost of these re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1507.00210","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":"1507.00210","created_at":"2026-05-18T01:37:29.561421+00:00"},{"alias_kind":"arxiv_version","alias_value":"1507.00210v1","created_at":"2026-05-18T01:37:29.561421+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1507.00210","created_at":"2026-05-18T01:37:29.561421+00:00"},{"alias_kind":"pith_short_12","alias_value":"7OSXHIMA5Q4H","created_at":"2026-05-18T12:29:10.953037+00:00"},{"alias_kind":"pith_short_16","alias_value":"7OSXHIMA5Q4H5XAH","created_at":"2026-05-18T12:29:10.953037+00:00"},{"alias_kind":"pith_short_8","alias_value":"7OSXHIMA","created_at":"2026-05-18T12:29:10.953037+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/7OSXHIMA5Q4H5XAHE4QMHRANHM","json":"https://pith.science/pith/7OSXHIMA5Q4H5XAHE4QMHRANHM.json","graph_json":"https://pith.science/api/pith-number/7OSXHIMA5Q4H5XAHE4QMHRANHM/graph.json","events_json":"https://pith.science/api/pith-number/7OSXHIMA5Q4H5XAHE4QMHRANHM/events.json","paper":"https://pith.science/paper/7OSXHIMA"},"agent_actions":{"view_html":"https://pith.science/pith/7OSXHIMA5Q4H5XAHE4QMHRANHM","download_json":"https://pith.science/pith/7OSXHIMA5Q4H5XAHE4QMHRANHM.json","view_paper":"https://pith.science/paper/7OSXHIMA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1507.00210&json=true","fetch_graph":"https://pith.science/api/pith-number/7OSXHIMA5Q4H5XAHE4QMHRANHM/graph.json","fetch_events":"https://pith.science/api/pith-number/7OSXHIMA5Q4H5XAHE4QMHRANHM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7OSXHIMA5Q4H5XAHE4QMHRANHM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7OSXHIMA5Q4H5XAHE4QMHRANHM/action/storage_attestation","attest_author":"https://pith.science/pith/7OSXHIMA5Q4H5XAHE4QMHRANHM/action/author_attestation","sign_citation":"https://pith.science/pith/7OSXHIMA5Q4H5XAHE4QMHRANHM/action/citation_signature","submit_replication":"https://pith.science/pith/7OSXHIMA5Q4H5XAHE4QMHRANHM/action/replication_record"}},"created_at":"2026-05-18T01:37:29.561421+00:00","updated_at":"2026-05-18T01:37:29.561421+00:00"}