{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:VRL5X4CJ5ULM6B4BW2TP5KSWZE","short_pith_number":"pith:VRL5X4CJ","canonical_record":{"source":{"id":"1810.09184","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2018-10-22T11:34:32Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d86d793af925f6dee094a9895cab666c8fed0bd03e4ed57266779f7986e12028","abstract_canon_sha256":"4245ffda447cac49c6856a49f0a685677ec7c2ae689145a0c2f640ad34efaa75"},"schema_version":"1.0"},"canonical_sha256":"ac57dbf049ed16cf0781b6a6feaa56c93131f17e98c9956d51b16033efef7c35","source":{"kind":"arxiv","id":"1810.09184","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1810.09184","created_at":"2026-05-18T00:02:40Z"},{"alias_kind":"arxiv_version","alias_value":"1810.09184v1","created_at":"2026-05-18T00:02:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.09184","created_at":"2026-05-18T00:02:40Z"},{"alias_kind":"pith_short_12","alias_value":"VRL5X4CJ5ULM","created_at":"2026-05-18T12:32:59Z"},{"alias_kind":"pith_short_16","alias_value":"VRL5X4CJ5ULM6B4B","created_at":"2026-05-18T12:32:59Z"},{"alias_kind":"pith_short_8","alias_value":"VRL5X4CJ","created_at":"2026-05-18T12:32:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:VRL5X4CJ5ULM6B4BW2TP5KSWZE","target":"record","payload":{"canonical_record":{"source":{"id":"1810.09184","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2018-10-22T11:34:32Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d86d793af925f6dee094a9895cab666c8fed0bd03e4ed57266779f7986e12028","abstract_canon_sha256":"4245ffda447cac49c6856a49f0a685677ec7c2ae689145a0c2f640ad34efaa75"},"schema_version":"1.0"},"canonical_sha256":"ac57dbf049ed16cf0781b6a6feaa56c93131f17e98c9956d51b16033efef7c35","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:02:40.843018Z","signature_b64":"uEp0ywzXdOe6WKuroag3LOAjG+Zae6GWcPvUrXJU//y9kzQeand+Z5r8PoUMf8o5MGz6wT5znZtATFmVzvrkAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ac57dbf049ed16cf0781b6a6feaa56c93131f17e98c9956d51b16033efef7c35","last_reissued_at":"2026-05-18T00:02:40.842528Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:02:40.842528Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1810.09184","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":"2026-05-18T00:02:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qccpdjA/9Tpt1iUlXaPGJ0iikpc+3MRbudcvjIy0dS9GBzbrn+ZUzamKgdVTYKvuK/sc8A0+8V80raaNpG6tCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T14:58:30.911060Z"},"content_sha256":"e9c7a89fd7bca893da77150d881ae031d6be476857bd12e1734b65d0303c76f1","schema_version":"1.0","event_id":"sha256:e9c7a89fd7bca893da77150d881ae031d6be476857bd12e1734b65d0303c76f1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:VRL5X4CJ5ULM6B4BW2TP5KSWZE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learning sparse transformations through backpropagation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Peter Bloem","submitted_at":"2018-10-22T11:34:32Z","abstract_excerpt":"Many transformations in deep learning architectures are sparsely connected. When such transformations cannot be designed by hand, they can be learned, even through plain backpropagation, for instance in attention mechanisms. However, during learning, such sparse structures are often represented in a dense form, as we do not know beforehand which elements will eventually become non-zero. We introduce the adaptive, sparse hyperlayer, a method for learning a sparse transformation, paramatrized sparsely: as index-tuples with associated values. To overcome the lack of gradients from such a discrete"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.09184","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":"2026-05-18T00:02:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"otNrVJV44APDaC5zFgusmFTYXfMNTU8KlLYqNQJ2tUK0KeOeOZaEJHQ29TQL9UgK6d7aOmz62XU1wUINt6WUAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T14:58:30.911418Z"},"content_sha256":"bbe1437a44ae3980255fd4846278e0428d71319f58cf9b2823de97bb026ed1ff","schema_version":"1.0","event_id":"sha256:bbe1437a44ae3980255fd4846278e0428d71319f58cf9b2823de97bb026ed1ff"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VRL5X4CJ5ULM6B4BW2TP5KSWZE/bundle.json","state_url":"https://pith.science/pith/VRL5X4CJ5ULM6B4BW2TP5KSWZE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VRL5X4CJ5ULM6B4BW2TP5KSWZE/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":"2026-06-27T14:58:30Z","links":{"resolver":"https://pith.science/pith/VRL5X4CJ5ULM6B4BW2TP5KSWZE","bundle":"https://pith.science/pith/VRL5X4CJ5ULM6B4BW2TP5KSWZE/bundle.json","state":"https://pith.science/pith/VRL5X4CJ5ULM6B4BW2TP5KSWZE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VRL5X4CJ5ULM6B4BW2TP5KSWZE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:VRL5X4CJ5ULM6B4BW2TP5KSWZE","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":"4245ffda447cac49c6856a49f0a685677ec7c2ae689145a0c2f640ad34efaa75","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2018-10-22T11:34:32Z","title_canon_sha256":"d86d793af925f6dee094a9895cab666c8fed0bd03e4ed57266779f7986e12028"},"schema_version":"1.0","source":{"id":"1810.09184","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1810.09184","created_at":"2026-05-18T00:02:40Z"},{"alias_kind":"arxiv_version","alias_value":"1810.09184v1","created_at":"2026-05-18T00:02:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.09184","created_at":"2026-05-18T00:02:40Z"},{"alias_kind":"pith_short_12","alias_value":"VRL5X4CJ5ULM","created_at":"2026-05-18T12:32:59Z"},{"alias_kind":"pith_short_16","alias_value":"VRL5X4CJ5ULM6B4B","created_at":"2026-05-18T12:32:59Z"},{"alias_kind":"pith_short_8","alias_value":"VRL5X4CJ","created_at":"2026-05-18T12:32:59Z"}],"graph_snapshots":[{"event_id":"sha256:bbe1437a44ae3980255fd4846278e0428d71319f58cf9b2823de97bb026ed1ff","target":"graph","created_at":"2026-05-18T00:02:40Z","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":"Many transformations in deep learning architectures are sparsely connected. When such transformations cannot be designed by hand, they can be learned, even through plain backpropagation, for instance in attention mechanisms. However, during learning, such sparse structures are often represented in a dense form, as we do not know beforehand which elements will eventually become non-zero. We introduce the adaptive, sparse hyperlayer, a method for learning a sparse transformation, paramatrized sparsely: as index-tuples with associated values. To overcome the lack of gradients from such a discrete","authors_text":"Peter Bloem","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2018-10-22T11:34:32Z","title":"Learning sparse transformations through backpropagation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.09184","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:e9c7a89fd7bca893da77150d881ae031d6be476857bd12e1734b65d0303c76f1","target":"record","created_at":"2026-05-18T00:02:40Z","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":"4245ffda447cac49c6856a49f0a685677ec7c2ae689145a0c2f640ad34efaa75","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2018-10-22T11:34:32Z","title_canon_sha256":"d86d793af925f6dee094a9895cab666c8fed0bd03e4ed57266779f7986e12028"},"schema_version":"1.0","source":{"id":"1810.09184","kind":"arxiv","version":1}},"canonical_sha256":"ac57dbf049ed16cf0781b6a6feaa56c93131f17e98c9956d51b16033efef7c35","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ac57dbf049ed16cf0781b6a6feaa56c93131f17e98c9956d51b16033efef7c35","first_computed_at":"2026-05-18T00:02:40.842528Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:02:40.842528Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uEp0ywzXdOe6WKuroag3LOAjG+Zae6GWcPvUrXJU//y9kzQeand+Z5r8PoUMf8o5MGz6wT5znZtATFmVzvrkAA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:02:40.843018Z","signed_message":"canonical_sha256_bytes"},"source_id":"1810.09184","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e9c7a89fd7bca893da77150d881ae031d6be476857bd12e1734b65d0303c76f1","sha256:bbe1437a44ae3980255fd4846278e0428d71319f58cf9b2823de97bb026ed1ff"],"state_sha256":"dfb65174a5ab36605fc363321c8a6504108ad07c5345bcbb19c93306f0ecbc63"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"R4/4WlwenZwkaG0zkyjaxRyDD+fUD68QnsxdRLobqnG1zbPFlndLnb+EhDP2SLLSK11fbA7c97VVDoXc7W5VBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-27T14:58:30.913347Z","bundle_sha256":"b09c2a2378a460d47f35d0f1d848a0b99e40f7d74c27df9ef95021007db30442"}}