{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:CZB2CJ7F7EFYPGG7N6MQEC7ONL","short_pith_number":"pith:CZB2CJ7F","canonical_record":{"source":{"id":"1705.07199","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-05-19T21:33:00Z","cross_cats_sorted":[],"title_canon_sha256":"706c12548957de49c4ed05dd2878dcb7a4d1d70feea17df6043d3f83d8526dc7","abstract_canon_sha256":"73467d7f30ecc28e27749f5472976efcdd938aa42974b31b5d8c8cdee1502e05"},"schema_version":"1.0"},"canonical_sha256":"1643a127e5f90b8798df6f99020bee6ae05b07983d9ef77df768da205c00240a","source":{"kind":"arxiv","id":"1705.07199","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1705.07199","created_at":"2026-05-18T00:44:06Z"},{"alias_kind":"arxiv_version","alias_value":"1705.07199v1","created_at":"2026-05-18T00:44:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1705.07199","created_at":"2026-05-18T00:44:06Z"},{"alias_kind":"pith_short_12","alias_value":"CZB2CJ7F7EFY","created_at":"2026-05-18T12:31:10Z"},{"alias_kind":"pith_short_16","alias_value":"CZB2CJ7F7EFYPGG7","created_at":"2026-05-18T12:31:10Z"},{"alias_kind":"pith_short_8","alias_value":"CZB2CJ7F","created_at":"2026-05-18T12:31:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:CZB2CJ7F7EFYPGG7N6MQEC7ONL","target":"record","payload":{"canonical_record":{"source":{"id":"1705.07199","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-05-19T21:33:00Z","cross_cats_sorted":[],"title_canon_sha256":"706c12548957de49c4ed05dd2878dcb7a4d1d70feea17df6043d3f83d8526dc7","abstract_canon_sha256":"73467d7f30ecc28e27749f5472976efcdd938aa42974b31b5d8c8cdee1502e05"},"schema_version":"1.0"},"canonical_sha256":"1643a127e5f90b8798df6f99020bee6ae05b07983d9ef77df768da205c00240a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:44:06.338110Z","signature_b64":"lWVO+eCMt11+Ok8g7/R7ONkPfsZVnWm5CeH7oAunCbPvNQNvFS2GUO5ZTcV9R4nsyHYQf3b0k4JARnt9M2/jCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1643a127e5f90b8798df6f99020bee6ae05b07983d9ef77df768da205c00240a","last_reissued_at":"2026-05-18T00:44:06.337450Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:44:06.337450Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1705.07199","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:44:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4F/SyR3GDh1bwUK2DUZdb+d52L30pNAHztkui9qc3xsW1+KEjgBtOcPipnrtITEYbs2oguSADhyWAWYjW67jBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T19:07:53.192292Z"},"content_sha256":"a4c4f1f235befe9fb04a3b0739434bec80b44e3fc8df17e205f4ea665264a5a6","schema_version":"1.0","event_id":"sha256:a4c4f1f235befe9fb04a3b0739434bec80b44e3fc8df17e205f4ea665264a5a6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:CZB2CJ7F7EFYPGG7N6MQEC7ONL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The High-Dimensional Geometry of Binary Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alexander G. Anderson, Cory P. Berg","submitted_at":"2017-05-19T21:33:00Z","abstract_excerpt":"Recent research has shown that one can train a neural network with binary weights and activations at train time by augmenting the weights with a high-precision continuous latent variable that accumulates small changes from stochastic gradient descent. However, there is a dearth of theoretical analysis to explain why we can effectively capture the features in our data with binary weights and activations. Our main result is that the neural networks with binary weights and activations trained using the method of Courbariaux, Hubara et al. (2016) work because of the high-dimensional geometry of bi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1705.07199","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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:44:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LcbY7/ak6BNxC7YjJ2bbjNYnT1r71KZ7gUa+ZaceKkXFMMOwtcgxDfEcAY4ZC5/rG+ykMYiKhv6+ruGCmFTTDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T19:07:53.192880Z"},"content_sha256":"8126fdb31f3f91d446df855864d7c5c5372f9f2535656a407e01dcdc79824b7a","schema_version":"1.0","event_id":"sha256:8126fdb31f3f91d446df855864d7c5c5372f9f2535656a407e01dcdc79824b7a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CZB2CJ7F7EFYPGG7N6MQEC7ONL/bundle.json","state_url":"https://pith.science/pith/CZB2CJ7F7EFYPGG7N6MQEC7ONL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CZB2CJ7F7EFYPGG7N6MQEC7ONL/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-06T19:07:53Z","links":{"resolver":"https://pith.science/pith/CZB2CJ7F7EFYPGG7N6MQEC7ONL","bundle":"https://pith.science/pith/CZB2CJ7F7EFYPGG7N6MQEC7ONL/bundle.json","state":"https://pith.science/pith/CZB2CJ7F7EFYPGG7N6MQEC7ONL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CZB2CJ7F7EFYPGG7N6MQEC7ONL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:CZB2CJ7F7EFYPGG7N6MQEC7ONL","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"73467d7f30ecc28e27749f5472976efcdd938aa42974b31b5d8c8cdee1502e05","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-05-19T21:33:00Z","title_canon_sha256":"706c12548957de49c4ed05dd2878dcb7a4d1d70feea17df6043d3f83d8526dc7"},"schema_version":"1.0","source":{"id":"1705.07199","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1705.07199","created_at":"2026-05-18T00:44:06Z"},{"alias_kind":"arxiv_version","alias_value":"1705.07199v1","created_at":"2026-05-18T00:44:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1705.07199","created_at":"2026-05-18T00:44:06Z"},{"alias_kind":"pith_short_12","alias_value":"CZB2CJ7F7EFY","created_at":"2026-05-18T12:31:10Z"},{"alias_kind":"pith_short_16","alias_value":"CZB2CJ7F7EFYPGG7","created_at":"2026-05-18T12:31:10Z"},{"alias_kind":"pith_short_8","alias_value":"CZB2CJ7F","created_at":"2026-05-18T12:31:10Z"}],"graph_snapshots":[{"event_id":"sha256:8126fdb31f3f91d446df855864d7c5c5372f9f2535656a407e01dcdc79824b7a","target":"graph","created_at":"2026-05-18T00:44:06Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Recent research has shown that one can train a neural network with binary weights and activations at train time by augmenting the weights with a high-precision continuous latent variable that accumulates small changes from stochastic gradient descent. However, there is a dearth of theoretical analysis to explain why we can effectively capture the features in our data with binary weights and activations. Our main result is that the neural networks with binary weights and activations trained using the method of Courbariaux, Hubara et al. (2016) work because of the high-dimensional geometry of bi","authors_text":"Alexander G. Anderson, Cory P. Berg","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-05-19T21:33:00Z","title":"The High-Dimensional Geometry of Binary Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1705.07199","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a4c4f1f235befe9fb04a3b0739434bec80b44e3fc8df17e205f4ea665264a5a6","target":"record","created_at":"2026-05-18T00:44:06Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"73467d7f30ecc28e27749f5472976efcdd938aa42974b31b5d8c8cdee1502e05","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-05-19T21:33:00Z","title_canon_sha256":"706c12548957de49c4ed05dd2878dcb7a4d1d70feea17df6043d3f83d8526dc7"},"schema_version":"1.0","source":{"id":"1705.07199","kind":"arxiv","version":1}},"canonical_sha256":"1643a127e5f90b8798df6f99020bee6ae05b07983d9ef77df768da205c00240a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1643a127e5f90b8798df6f99020bee6ae05b07983d9ef77df768da205c00240a","first_computed_at":"2026-05-18T00:44:06.337450Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:44:06.337450Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lWVO+eCMt11+Ok8g7/R7ONkPfsZVnWm5CeH7oAunCbPvNQNvFS2GUO5ZTcV9R4nsyHYQf3b0k4JARnt9M2/jCQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:44:06.338110Z","signed_message":"canonical_sha256_bytes"},"source_id":"1705.07199","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a4c4f1f235befe9fb04a3b0739434bec80b44e3fc8df17e205f4ea665264a5a6","sha256:8126fdb31f3f91d446df855864d7c5c5372f9f2535656a407e01dcdc79824b7a"],"state_sha256":"9db56d29a498b2a2cd4fa93e43fa18a8fa2d672f7189c6820e12009170e7a5f6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5BZyqUQ0LqOPsgKKd/QYDViWwnnC+cnW8aGnJm6E2NHVQbfm79OeIMNtigcvOCa3GOmIyS4kuEKnVtW2MhKJAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-06T19:07:53.196677Z","bundle_sha256":"53c428b97c0b3ab383f44c239d3ac08b9691e24faae0c874e721bc43056dc7e9"}}