{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:Q7NXNL5XRCREVPMKSURTST4Q3U","short_pith_number":"pith:Q7NXNL5X","canonical_record":{"source":{"id":"1702.00071","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-01-31T22:14:59Z","cross_cats_sorted":["cs.NE"],"title_canon_sha256":"442df78a25e477b392a61aa6d5b5785c5a3093ca2f7494a809413c0e47da81d8","abstract_canon_sha256":"076791c1459153c7796af247e6e0ffe66fbfa98ac6eec572bc6ff01f1ae64eef"},"schema_version":"1.0"},"canonical_sha256":"87db76afb788a24abd8a9523394f90dd2de5a394f31be1d0e3b2a5a34494d42c","source":{"kind":"arxiv","id":"1702.00071","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1702.00071","created_at":"2026-05-18T00:33:02Z"},{"alias_kind":"arxiv_version","alias_value":"1702.00071v4","created_at":"2026-05-18T00:33:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1702.00071","created_at":"2026-05-18T00:33:02Z"},{"alias_kind":"pith_short_12","alias_value":"Q7NXNL5XRCRE","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_16","alias_value":"Q7NXNL5XRCREVPMK","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_8","alias_value":"Q7NXNL5X","created_at":"2026-05-18T12:31:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:Q7NXNL5XRCREVPMKSURTST4Q3U","target":"record","payload":{"canonical_record":{"source":{"id":"1702.00071","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-01-31T22:14:59Z","cross_cats_sorted":["cs.NE"],"title_canon_sha256":"442df78a25e477b392a61aa6d5b5785c5a3093ca2f7494a809413c0e47da81d8","abstract_canon_sha256":"076791c1459153c7796af247e6e0ffe66fbfa98ac6eec572bc6ff01f1ae64eef"},"schema_version":"1.0"},"canonical_sha256":"87db76afb788a24abd8a9523394f90dd2de5a394f31be1d0e3b2a5a34494d42c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:33:02.235799Z","signature_b64":"cTMJj2m21+3eTuJf/HN7w9B7IWDirSvomGtQ7XZBzkD9jWATc9DSlG2Cgak1zBHTh3RIWFJrRbC1Mp9msgKKBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"87db76afb788a24abd8a9523394f90dd2de5a394f31be1d0e3b2a5a34494d42c","last_reissued_at":"2026-05-18T00:33:02.235310Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:33:02.235310Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1702.00071","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:33:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dcn2BMy2FB8+fEtvS5+DLzRhSr6zIGJWu7tkf4j6G0G2pIZ2KSzGm2haaYuBWR0vXDa7jNP8lLAQzDE0xiplCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T04:14:21.912273Z"},"content_sha256":"09b9c6456f188436a13082228a8bac2242f0d1c54571ad58acec16dbabd0f983","schema_version":"1.0","event_id":"sha256:09b9c6456f188436a13082228a8bac2242f0d1c54571ad58acec16dbabd0f983"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:Q7NXNL5XRCREVPMKSURTST4Q3U","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On orthogonality and learning recurrent networks with long term dependencies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE"],"primary_cat":"cs.LG","authors_text":"Chiheb Trabelsi, Chris Pal, Eugene Vorontsov, Samuel Kadoury","submitted_at":"2017-01-31T22:14:59Z","abstract_excerpt":"It is well known that it is challenging to train deep neural networks and recurrent neural networks for tasks that exhibit long term dependencies. The vanishing or exploding gradient problem is a well known issue associated with these challenges. One approach to addressing vanishing and exploding gradients is to use either soft or hard constraints on weight matrices so as to encourage or enforce orthogonality. Orthogonal matrices preserve gradient norm during backpropagation and may therefore be a desirable property. This paper explores issues with optimization convergence, speed and gradient "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1702.00071","kind":"arxiv","version":4},"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":"2026-05-18T00:33:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CXNqdCud2U49UU73VAJI9iox+4d1WNNa9TcCEWSCjPpjslb6N2vcm+8qBtOisHSs4oY5dKdZNLL/po7XGzfWAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T04:14:21.912666Z"},"content_sha256":"3af2bd271df10b7261afd992a00a1bf580c81c6d681113506290b4559dea551f","schema_version":"1.0","event_id":"sha256:3af2bd271df10b7261afd992a00a1bf580c81c6d681113506290b4559dea551f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Q7NXNL5XRCREVPMKSURTST4Q3U/bundle.json","state_url":"https://pith.science/pith/Q7NXNL5XRCREVPMKSURTST4Q3U/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Q7NXNL5XRCREVPMKSURTST4Q3U/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":"2026-05-26T04:14:21Z","links":{"resolver":"https://pith.science/pith/Q7NXNL5XRCREVPMKSURTST4Q3U","bundle":"https://pith.science/pith/Q7NXNL5XRCREVPMKSURTST4Q3U/bundle.json","state":"https://pith.science/pith/Q7NXNL5XRCREVPMKSURTST4Q3U/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Q7NXNL5XRCREVPMKSURTST4Q3U/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:Q7NXNL5XRCREVPMKSURTST4Q3U","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":"076791c1459153c7796af247e6e0ffe66fbfa98ac6eec572bc6ff01f1ae64eef","cross_cats_sorted":["cs.NE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-01-31T22:14:59Z","title_canon_sha256":"442df78a25e477b392a61aa6d5b5785c5a3093ca2f7494a809413c0e47da81d8"},"schema_version":"1.0","source":{"id":"1702.00071","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1702.00071","created_at":"2026-05-18T00:33:02Z"},{"alias_kind":"arxiv_version","alias_value":"1702.00071v4","created_at":"2026-05-18T00:33:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1702.00071","created_at":"2026-05-18T00:33:02Z"},{"alias_kind":"pith_short_12","alias_value":"Q7NXNL5XRCRE","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_16","alias_value":"Q7NXNL5XRCREVPMK","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_8","alias_value":"Q7NXNL5X","created_at":"2026-05-18T12:31:37Z"}],"graph_snapshots":[{"event_id":"sha256:3af2bd271df10b7261afd992a00a1bf580c81c6d681113506290b4559dea551f","target":"graph","created_at":"2026-05-18T00:33:02Z","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":"It is well known that it is challenging to train deep neural networks and recurrent neural networks for tasks that exhibit long term dependencies. The vanishing or exploding gradient problem is a well known issue associated with these challenges. One approach to addressing vanishing and exploding gradients is to use either soft or hard constraints on weight matrices so as to encourage or enforce orthogonality. Orthogonal matrices preserve gradient norm during backpropagation and may therefore be a desirable property. This paper explores issues with optimization convergence, speed and gradient ","authors_text":"Chiheb Trabelsi, Chris Pal, Eugene Vorontsov, Samuel Kadoury","cross_cats":["cs.NE"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-01-31T22:14:59Z","title":"On orthogonality and learning recurrent networks with long term dependencies"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1702.00071","kind":"arxiv","version":4},"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":"sha256:09b9c6456f188436a13082228a8bac2242f0d1c54571ad58acec16dbabd0f983","target":"record","created_at":"2026-05-18T00:33:02Z","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":"076791c1459153c7796af247e6e0ffe66fbfa98ac6eec572bc6ff01f1ae64eef","cross_cats_sorted":["cs.NE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-01-31T22:14:59Z","title_canon_sha256":"442df78a25e477b392a61aa6d5b5785c5a3093ca2f7494a809413c0e47da81d8"},"schema_version":"1.0","source":{"id":"1702.00071","kind":"arxiv","version":4}},"canonical_sha256":"87db76afb788a24abd8a9523394f90dd2de5a394f31be1d0e3b2a5a34494d42c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"87db76afb788a24abd8a9523394f90dd2de5a394f31be1d0e3b2a5a34494d42c","first_computed_at":"2026-05-18T00:33:02.235310Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:33:02.235310Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cTMJj2m21+3eTuJf/HN7w9B7IWDirSvomGtQ7XZBzkD9jWATc9DSlG2Cgak1zBHTh3RIWFJrRbC1Mp9msgKKBQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:33:02.235799Z","signed_message":"canonical_sha256_bytes"},"source_id":"1702.00071","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:09b9c6456f188436a13082228a8bac2242f0d1c54571ad58acec16dbabd0f983","sha256:3af2bd271df10b7261afd992a00a1bf580c81c6d681113506290b4559dea551f"],"state_sha256":"f8ac44ea75b8f2848bc0315ca1d49082879d05fee37652b7b2bf1af2cace39cf"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WUvQbBv8xDAkmXr0v2l5a0sy5rlMC1zj8guXLfNOZJgbOl7QGVKO2p7LDoBoJ2sp8N2gXyedMzaYbWNCPY0+AA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T04:14:21.915707Z","bundle_sha256":"ac79134c41080a09d8e22d6dbb2e16229b634b14ec240ac73c26b6104d985d8a"}}