{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:FK6JQG4LG2GUYBMAXFTCWK6YHJ","short_pith_number":"pith:FK6JQG4L","canonical_record":{"source":{"id":"1708.08863","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-08-29T16:18:44Z","cross_cats_sorted":["cs.IT","cs.LG","math.IT"],"title_canon_sha256":"31351208afc3e47aba94753592d9be24c171df0763ca6537fde766eae8dcc9fe","abstract_canon_sha256":"2b0fb30f9887a49199a78dcd0b1c138b89fff6966f8352a932e17c4eacfc7415"},"schema_version":"1.0"},"canonical_sha256":"2abc981b8b368d4c0580b9662b2bd83a6297e9d3cdaff6646ea61c0aa7db944b","source":{"kind":"arxiv","id":"1708.08863","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1708.08863","created_at":"2026-05-18T00:15:10Z"},{"alias_kind":"arxiv_version","alias_value":"1708.08863v2","created_at":"2026-05-18T00:15:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1708.08863","created_at":"2026-05-18T00:15:10Z"},{"alias_kind":"pith_short_12","alias_value":"FK6JQG4LG2GU","created_at":"2026-05-18T12:31:15Z"},{"alias_kind":"pith_short_16","alias_value":"FK6JQG4LG2GUYBMA","created_at":"2026-05-18T12:31:15Z"},{"alias_kind":"pith_short_8","alias_value":"FK6JQG4L","created_at":"2026-05-18T12:31:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:FK6JQG4LG2GUYBMAXFTCWK6YHJ","target":"record","payload":{"canonical_record":{"source":{"id":"1708.08863","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-08-29T16:18:44Z","cross_cats_sorted":["cs.IT","cs.LG","math.IT"],"title_canon_sha256":"31351208afc3e47aba94753592d9be24c171df0763ca6537fde766eae8dcc9fe","abstract_canon_sha256":"2b0fb30f9887a49199a78dcd0b1c138b89fff6966f8352a932e17c4eacfc7415"},"schema_version":"1.0"},"canonical_sha256":"2abc981b8b368d4c0580b9662b2bd83a6297e9d3cdaff6646ea61c0aa7db944b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:15:10.261357Z","signature_b64":"ajRxWNqrKD2En4/ka1btLgYNXh7q3EffYemSLaif9SqAjY798J1ivb3R2ku0f+Umf4G37LFJwsH9NIktkFVpCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2abc981b8b368d4c0580b9662b2bd83a6297e9d3cdaff6646ea61c0aa7db944b","last_reissued_at":"2026-05-18T00:15:10.260602Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:15:10.260602Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1708.08863","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:15:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hcgRX9ixNtMZRdpaRz440zri2CK3/RvMMhDV0r0mYwfHZnxNAw11Oa7tXYeo8/pGngKGrxJ+rYjvVZoapSqyBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T22:34:33.891477Z"},"content_sha256":"5ddb0e6958e37f7d7568a06022c0ac4a7801408da1c4d3107cb4fa1aee44e2e7","schema_version":"1.0","event_id":"sha256:5ddb0e6958e37f7d7568a06022c0ac4a7801408da1c4d3107cb4fa1aee44e2e7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:FK6JQG4LG2GUYBMAXFTCWK6YHJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Gradual Learning of Recurrent Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IT","cs.LG","math.IT"],"primary_cat":"stat.ML","authors_text":"Gal Rattner, Haim Permuter, Ziv Aharoni","submitted_at":"2017-08-29T16:18:44Z","abstract_excerpt":"Recurrent Neural Networks (RNNs) achieve state-of-the-art results in many sequence-to-sequence modeling tasks. However, RNNs are difficult to train and tend to suffer from overfitting. Motivated by the Data Processing Inequality (DPI), we formulate the multi-layered network as a Markov chain, introducing a training method that comprises training the network gradually and using layer-wise gradient clipping. We found that applying our methods, combined with previously introduced regularization and optimization methods, resulted in improvements in state-of-the-art architectures operating in langu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1708.08863","kind":"arxiv","version":2},"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":"2026-05-18T00:15:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TkAUU0aZeC8AFqcQ9XfHi7rIXlpZLxn2XzRmHMhtFm+Lcd2VRP5xD5P9c3vqdEEGg8nQzi6ZbAyOZN27niF2Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T22:34:33.891835Z"},"content_sha256":"f4af9b82e761f3a35bb6320845096462953201a9d2553fae99c72650114e3efc","schema_version":"1.0","event_id":"sha256:f4af9b82e761f3a35bb6320845096462953201a9d2553fae99c72650114e3efc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FK6JQG4LG2GUYBMAXFTCWK6YHJ/bundle.json","state_url":"https://pith.science/pith/FK6JQG4LG2GUYBMAXFTCWK6YHJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FK6JQG4LG2GUYBMAXFTCWK6YHJ/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":"2026-05-31T22:34:33Z","links":{"resolver":"https://pith.science/pith/FK6JQG4LG2GUYBMAXFTCWK6YHJ","bundle":"https://pith.science/pith/FK6JQG4LG2GUYBMAXFTCWK6YHJ/bundle.json","state":"https://pith.science/pith/FK6JQG4LG2GUYBMAXFTCWK6YHJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FK6JQG4LG2GUYBMAXFTCWK6YHJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:FK6JQG4LG2GUYBMAXFTCWK6YHJ","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":"2b0fb30f9887a49199a78dcd0b1c138b89fff6966f8352a932e17c4eacfc7415","cross_cats_sorted":["cs.IT","cs.LG","math.IT"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-08-29T16:18:44Z","title_canon_sha256":"31351208afc3e47aba94753592d9be24c171df0763ca6537fde766eae8dcc9fe"},"schema_version":"1.0","source":{"id":"1708.08863","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1708.08863","created_at":"2026-05-18T00:15:10Z"},{"alias_kind":"arxiv_version","alias_value":"1708.08863v2","created_at":"2026-05-18T00:15:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1708.08863","created_at":"2026-05-18T00:15:10Z"},{"alias_kind":"pith_short_12","alias_value":"FK6JQG4LG2GU","created_at":"2026-05-18T12:31:15Z"},{"alias_kind":"pith_short_16","alias_value":"FK6JQG4LG2GUYBMA","created_at":"2026-05-18T12:31:15Z"},{"alias_kind":"pith_short_8","alias_value":"FK6JQG4L","created_at":"2026-05-18T12:31:15Z"}],"graph_snapshots":[{"event_id":"sha256:f4af9b82e761f3a35bb6320845096462953201a9d2553fae99c72650114e3efc","target":"graph","created_at":"2026-05-18T00:15:10Z","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":"Recurrent Neural Networks (RNNs) achieve state-of-the-art results in many sequence-to-sequence modeling tasks. However, RNNs are difficult to train and tend to suffer from overfitting. Motivated by the Data Processing Inequality (DPI), we formulate the multi-layered network as a Markov chain, introducing a training method that comprises training the network gradually and using layer-wise gradient clipping. We found that applying our methods, combined with previously introduced regularization and optimization methods, resulted in improvements in state-of-the-art architectures operating in langu","authors_text":"Gal Rattner, Haim Permuter, Ziv Aharoni","cross_cats":["cs.IT","cs.LG","math.IT"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-08-29T16:18:44Z","title":"Gradual Learning of Recurrent Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1708.08863","kind":"arxiv","version":2},"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":"sha256:5ddb0e6958e37f7d7568a06022c0ac4a7801408da1c4d3107cb4fa1aee44e2e7","target":"record","created_at":"2026-05-18T00:15:10Z","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":"2b0fb30f9887a49199a78dcd0b1c138b89fff6966f8352a932e17c4eacfc7415","cross_cats_sorted":["cs.IT","cs.LG","math.IT"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-08-29T16:18:44Z","title_canon_sha256":"31351208afc3e47aba94753592d9be24c171df0763ca6537fde766eae8dcc9fe"},"schema_version":"1.0","source":{"id":"1708.08863","kind":"arxiv","version":2}},"canonical_sha256":"2abc981b8b368d4c0580b9662b2bd83a6297e9d3cdaff6646ea61c0aa7db944b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2abc981b8b368d4c0580b9662b2bd83a6297e9d3cdaff6646ea61c0aa7db944b","first_computed_at":"2026-05-18T00:15:10.260602Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:15:10.260602Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ajRxWNqrKD2En4/ka1btLgYNXh7q3EffYemSLaif9SqAjY798J1ivb3R2ku0f+Umf4G37LFJwsH9NIktkFVpCA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:15:10.261357Z","signed_message":"canonical_sha256_bytes"},"source_id":"1708.08863","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5ddb0e6958e37f7d7568a06022c0ac4a7801408da1c4d3107cb4fa1aee44e2e7","sha256:f4af9b82e761f3a35bb6320845096462953201a9d2553fae99c72650114e3efc"],"state_sha256":"3c05197f4120e1e77c2cf9bad1902311c0884bff391126e0e3d08c855868ec08"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HPfUzqPNnwJiqHSHHm25JpNq4Vj/GZ3IFpCK/b21ZgCH55scDwBlkEgcWCRB6y4Ztwc1SQKiv2vI3Bzd3AKiAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T22:34:33.894283Z","bundle_sha256":"7eeefce477a7b009dc48cb62c427f3bde3b2557b46417ea9097de5b5820e8246"}}