{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:UVSFR7NC2BQGPDYN42RJODK54K","short_pith_number":"pith:UVSFR7NC","canonical_record":{"source":{"id":"1808.05285","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-08-15T21:13:06Z","cross_cats_sorted":[],"title_canon_sha256":"1c600d483540a4dbec25fa73615c6bc80334b8fc9aa5e3a85c8db26e49c0f0ad","abstract_canon_sha256":"5956242fe0ca07b0fe2e8f81a8caf455203755eb0a4cb20d72f0c4cbb20e83de"},"schema_version":"1.0"},"canonical_sha256":"a56458fda2d060678f0de6a2970d5de2a63dc5c1ecdec52db75175704a802329","source":{"kind":"arxiv","id":"1808.05285","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1808.05285","created_at":"2026-05-18T00:07:57Z"},{"alias_kind":"arxiv_version","alias_value":"1808.05285v1","created_at":"2026-05-18T00:07:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1808.05285","created_at":"2026-05-18T00:07:57Z"},{"alias_kind":"pith_short_12","alias_value":"UVSFR7NC2BQG","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_16","alias_value":"UVSFR7NC2BQGPDYN","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_8","alias_value":"UVSFR7NC","created_at":"2026-05-18T12:32:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:UVSFR7NC2BQGPDYN42RJODK54K","target":"record","payload":{"canonical_record":{"source":{"id":"1808.05285","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-08-15T21:13:06Z","cross_cats_sorted":[],"title_canon_sha256":"1c600d483540a4dbec25fa73615c6bc80334b8fc9aa5e3a85c8db26e49c0f0ad","abstract_canon_sha256":"5956242fe0ca07b0fe2e8f81a8caf455203755eb0a4cb20d72f0c4cbb20e83de"},"schema_version":"1.0"},"canonical_sha256":"a56458fda2d060678f0de6a2970d5de2a63dc5c1ecdec52db75175704a802329","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:07:57.645309Z","signature_b64":"inqH3PZNfZR3DrD94Nw2wINTLejk80UbMpAQTVT1A3bj9HiVpUgSKRxlyuih22uRbZxDZVw/hd3CWkYWiQ7QDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a56458fda2d060678f0de6a2970d5de2a63dc5c1ecdec52db75175704a802329","last_reissued_at":"2026-05-18T00:07:57.644590Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:07:57.644590Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1808.05285","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":"2026-05-18T00:07:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5ELxDzp6U3VfPB/3fPWov+bwnEIU3xehP/75kLgXzgWwf3GhGI6K+xiVFobSbNZ3kdn1uNvA10YyU3AMJOhLCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T19:11:42.475537Z"},"content_sha256":"d7a41f971b5da7d09eac6d2dbcd0df1540447a1a86c760fbe47d0d00a3c041be","schema_version":"1.0","event_id":"sha256:d7a41f971b5da7d09eac6d2dbcd0df1540447a1a86c760fbe47d0d00a3c041be"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:UVSFR7NC2BQGPDYN42RJODK54K","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DNN Feature Map Compression using Learned Representation over GF(2)","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alec Hodgkinson, Denis A. Gudovskiy, Luca Rigazio","submitted_at":"2018-08-15T21:13:06Z","abstract_excerpt":"In this paper, we introduce a method to compress intermediate feature maps of deep neural networks (DNNs) to decrease memory storage and bandwidth requirements during inference. Unlike previous works, the proposed method is based on converting fixed-point activations into vectors over the smallest GF(2) finite field followed by nonlinear dimensionality reduction (NDR) layers embedded into a DNN. Such an end-to-end learned representation finds more compact feature maps by exploiting quantization redundancies within the fixed-point activations along the channel or spatial dimensions. We apply th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1808.05285","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":"2026-05-18T00:07:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jWaml46Y+4ed1vlgLvIVnzzdI0udUzqdy1uDMU5VVKrsL4ENIs/0gtkkf6idPLK2PcYcRZ9E6J/DIvGJheewBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T19:11:42.475932Z"},"content_sha256":"31c23ea2713daee44682a42381c55b54f4a348747022c675f9107093668917ef","schema_version":"1.0","event_id":"sha256:31c23ea2713daee44682a42381c55b54f4a348747022c675f9107093668917ef"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UVSFR7NC2BQGPDYN42RJODK54K/bundle.json","state_url":"https://pith.science/pith/UVSFR7NC2BQGPDYN42RJODK54K/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UVSFR7NC2BQGPDYN42RJODK54K/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":"2026-05-26T19:11:42Z","links":{"resolver":"https://pith.science/pith/UVSFR7NC2BQGPDYN42RJODK54K","bundle":"https://pith.science/pith/UVSFR7NC2BQGPDYN42RJODK54K/bundle.json","state":"https://pith.science/pith/UVSFR7NC2BQGPDYN42RJODK54K/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UVSFR7NC2BQGPDYN42RJODK54K/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:UVSFR7NC2BQGPDYN42RJODK54K","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":"5956242fe0ca07b0fe2e8f81a8caf455203755eb0a4cb20d72f0c4cbb20e83de","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-08-15T21:13:06Z","title_canon_sha256":"1c600d483540a4dbec25fa73615c6bc80334b8fc9aa5e3a85c8db26e49c0f0ad"},"schema_version":"1.0","source":{"id":"1808.05285","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1808.05285","created_at":"2026-05-18T00:07:57Z"},{"alias_kind":"arxiv_version","alias_value":"1808.05285v1","created_at":"2026-05-18T00:07:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1808.05285","created_at":"2026-05-18T00:07:57Z"},{"alias_kind":"pith_short_12","alias_value":"UVSFR7NC2BQG","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_16","alias_value":"UVSFR7NC2BQGPDYN","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_8","alias_value":"UVSFR7NC","created_at":"2026-05-18T12:32:56Z"}],"graph_snapshots":[{"event_id":"sha256:31c23ea2713daee44682a42381c55b54f4a348747022c675f9107093668917ef","target":"graph","created_at":"2026-05-18T00:07:57Z","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":"In this paper, we introduce a method to compress intermediate feature maps of deep neural networks (DNNs) to decrease memory storage and bandwidth requirements during inference. Unlike previous works, the proposed method is based on converting fixed-point activations into vectors over the smallest GF(2) finite field followed by nonlinear dimensionality reduction (NDR) layers embedded into a DNN. Such an end-to-end learned representation finds more compact feature maps by exploiting quantization redundancies within the fixed-point activations along the channel or spatial dimensions. We apply th","authors_text":"Alec Hodgkinson, Denis A. Gudovskiy, Luca Rigazio","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-08-15T21:13:06Z","title":"DNN Feature Map Compression using Learned Representation over GF(2)"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1808.05285","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:d7a41f971b5da7d09eac6d2dbcd0df1540447a1a86c760fbe47d0d00a3c041be","target":"record","created_at":"2026-05-18T00:07:57Z","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":"5956242fe0ca07b0fe2e8f81a8caf455203755eb0a4cb20d72f0c4cbb20e83de","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-08-15T21:13:06Z","title_canon_sha256":"1c600d483540a4dbec25fa73615c6bc80334b8fc9aa5e3a85c8db26e49c0f0ad"},"schema_version":"1.0","source":{"id":"1808.05285","kind":"arxiv","version":1}},"canonical_sha256":"a56458fda2d060678f0de6a2970d5de2a63dc5c1ecdec52db75175704a802329","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a56458fda2d060678f0de6a2970d5de2a63dc5c1ecdec52db75175704a802329","first_computed_at":"2026-05-18T00:07:57.644590Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:07:57.644590Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"inqH3PZNfZR3DrD94Nw2wINTLejk80UbMpAQTVT1A3bj9HiVpUgSKRxlyuih22uRbZxDZVw/hd3CWkYWiQ7QDA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:07:57.645309Z","signed_message":"canonical_sha256_bytes"},"source_id":"1808.05285","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d7a41f971b5da7d09eac6d2dbcd0df1540447a1a86c760fbe47d0d00a3c041be","sha256:31c23ea2713daee44682a42381c55b54f4a348747022c675f9107093668917ef"],"state_sha256":"af47d6baf68ad60432de822255f279b5b6a7342dcc701da9e15854af01cf5342"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EZGfdfkiWL2LP4pjFLE8UAGo2WB+NHcYji+U4WSKVoyCPtmvwa9Qr6ObG5CqEobc+Z0egAQfhHfOZZBtkWsxDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T19:11:42.479497Z","bundle_sha256":"db5a486819c055dbad49dad8a50817564dc38fd54f89c8c5bb9a3c67d8b0fdfe"}}