{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:7FLUHHWVQEVBL3WOPHP6LHUFKF","short_pith_number":"pith:7FLUHHWV","canonical_record":{"source":{"id":"2004.14070","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-04-29T10:47:54Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"094187039008547022d5bf14f4d44832e8a953eebee3d0939b456e8ee38e3da0","abstract_canon_sha256":"d52616f56eae5bce366feda87ef4108afa25b50d56191cf15cb18fb9976ae279"},"schema_version":"1.0"},"canonical_sha256":"f957439ed5812a15eece79dfe59e85517c5d3d8e54ed5664e0ff4270e07ad611","source":{"kind":"arxiv","id":"2004.14070","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2004.14070","created_at":"2026-07-05T02:05:25Z"},{"alias_kind":"arxiv_version","alias_value":"2004.14070v4","created_at":"2026-07-05T02:05:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2004.14070","created_at":"2026-07-05T02:05:25Z"},{"alias_kind":"pith_short_12","alias_value":"7FLUHHWVQEVB","created_at":"2026-07-05T02:05:25Z"},{"alias_kind":"pith_short_16","alias_value":"7FLUHHWVQEVBL3WO","created_at":"2026-07-05T02:05:25Z"},{"alias_kind":"pith_short_8","alias_value":"7FLUHHWV","created_at":"2026-07-05T02:05:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:7FLUHHWVQEVBL3WOPHP6LHUFKF","target":"record","payload":{"canonical_record":{"source":{"id":"2004.14070","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-04-29T10:47:54Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"094187039008547022d5bf14f4d44832e8a953eebee3d0939b456e8ee38e3da0","abstract_canon_sha256":"d52616f56eae5bce366feda87ef4108afa25b50d56191cf15cb18fb9976ae279"},"schema_version":"1.0"},"canonical_sha256":"f957439ed5812a15eece79dfe59e85517c5d3d8e54ed5664e0ff4270e07ad611","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:05:25.147201Z","signature_b64":"qtK3o+ImQHlLuELe0hAVUY/PB8tO1ocTW+9xa5R89iFHSM0wHWZwvWRwcPgFjhyHyY5eN12hjfxl2ZtN0oq1AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f957439ed5812a15eece79dfe59e85517c5d3d8e54ed5664e0ff4270e07ad611","last_reissued_at":"2026-07-05T02:05:25.146719Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:05:25.146719Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2004.14070","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":"2026-07-05T02:05:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VYdFFVFr+D1BEZ7Niyal2yaVlo6USEkeuoI0pwXb2Xc/3UlXdp7Uz33PZ8OztPZqu7Nw5wpA1RAGKqO28VluBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T21:44:38.184367Z"},"content_sha256":"ecb233964a531abc991592ad950e49d57740640e5aaf3f5fb50f9c1a93be0eba","schema_version":"1.0","event_id":"sha256:ecb233964a531abc991592ad950e49d57740640e5aaf3f5fb50f9c1a93be0eba"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:7FLUHHWVQEVBL3WOPHP6LHUFKF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Continual Deep Learning by Functional Regularisation of Memorable Past","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Alexander Immer, Mohammad Emtiyaz Khan, Pingbo Pan, Richard E. Turner, Runa Eschenhagen, Siddharth Swaroop","submitted_at":"2020-04-29T10:47:54Z","abstract_excerpt":"Continually learning new skills is important for intelligent systems, yet standard deep learning methods suffer from catastrophic forgetting of the past. Recent works address this with weight regularisation. Functional regularisation, although computationally expensive, is expected to perform better, but rarely does so in practice. In this paper, we fix this issue by using a new functional-regularisation approach that utilises a few memorable past examples crucial to avoid forgetting. By using a Gaussian Process formulation of deep networks, our approach enables training in weight-space while "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2004.14070","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2004.14070/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":"2026-07-05T02:05:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CaF2FwR3X6k5h/PEGOQz9jbvi2u+xB1Te8weFVTRI+WBTEjQszXm4SeA9UnZ5pB+RDgq9HYfRx6iOlHB0k7yAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T21:44:38.184884Z"},"content_sha256":"f302fe518b5cd1b8ed33cce69f6184dd68ed7a58116a05ae030b11a8769c0b75","schema_version":"1.0","event_id":"sha256:f302fe518b5cd1b8ed33cce69f6184dd68ed7a58116a05ae030b11a8769c0b75"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7FLUHHWVQEVBL3WOPHP6LHUFKF/bundle.json","state_url":"https://pith.science/pith/7FLUHHWVQEVBL3WOPHP6LHUFKF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7FLUHHWVQEVBL3WOPHP6LHUFKF/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":"2026-08-03T21:44:38Z","links":{"resolver":"https://pith.science/pith/7FLUHHWVQEVBL3WOPHP6LHUFKF","bundle":"https://pith.science/pith/7FLUHHWVQEVBL3WOPHP6LHUFKF/bundle.json","state":"https://pith.science/pith/7FLUHHWVQEVBL3WOPHP6LHUFKF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7FLUHHWVQEVBL3WOPHP6LHUFKF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:7FLUHHWVQEVBL3WOPHP6LHUFKF","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":"d52616f56eae5bce366feda87ef4108afa25b50d56191cf15cb18fb9976ae279","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-04-29T10:47:54Z","title_canon_sha256":"094187039008547022d5bf14f4d44832e8a953eebee3d0939b456e8ee38e3da0"},"schema_version":"1.0","source":{"id":"2004.14070","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2004.14070","created_at":"2026-07-05T02:05:25Z"},{"alias_kind":"arxiv_version","alias_value":"2004.14070v4","created_at":"2026-07-05T02:05:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2004.14070","created_at":"2026-07-05T02:05:25Z"},{"alias_kind":"pith_short_12","alias_value":"7FLUHHWVQEVB","created_at":"2026-07-05T02:05:25Z"},{"alias_kind":"pith_short_16","alias_value":"7FLUHHWVQEVBL3WO","created_at":"2026-07-05T02:05:25Z"},{"alias_kind":"pith_short_8","alias_value":"7FLUHHWV","created_at":"2026-07-05T02:05:25Z"}],"graph_snapshots":[{"event_id":"sha256:f302fe518b5cd1b8ed33cce69f6184dd68ed7a58116a05ae030b11a8769c0b75","target":"graph","created_at":"2026-07-05T02:05:25Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2004.14070/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Continually learning new skills is important for intelligent systems, yet standard deep learning methods suffer from catastrophic forgetting of the past. Recent works address this with weight regularisation. Functional regularisation, although computationally expensive, is expected to perform better, but rarely does so in practice. In this paper, we fix this issue by using a new functional-regularisation approach that utilises a few memorable past examples crucial to avoid forgetting. By using a Gaussian Process formulation of deep networks, our approach enables training in weight-space while ","authors_text":"Alexander Immer, Mohammad Emtiyaz Khan, Pingbo Pan, Richard E. Turner, Runa Eschenhagen, Siddharth Swaroop","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-04-29T10:47:54Z","title":"Continual Deep Learning by Functional Regularisation of Memorable Past"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2004.14070","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:ecb233964a531abc991592ad950e49d57740640e5aaf3f5fb50f9c1a93be0eba","target":"record","created_at":"2026-07-05T02:05:25Z","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":"d52616f56eae5bce366feda87ef4108afa25b50d56191cf15cb18fb9976ae279","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-04-29T10:47:54Z","title_canon_sha256":"094187039008547022d5bf14f4d44832e8a953eebee3d0939b456e8ee38e3da0"},"schema_version":"1.0","source":{"id":"2004.14070","kind":"arxiv","version":4}},"canonical_sha256":"f957439ed5812a15eece79dfe59e85517c5d3d8e54ed5664e0ff4270e07ad611","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f957439ed5812a15eece79dfe59e85517c5d3d8e54ed5664e0ff4270e07ad611","first_computed_at":"2026-07-05T02:05:25.146719Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:05:25.146719Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qtK3o+ImQHlLuELe0hAVUY/PB8tO1ocTW+9xa5R89iFHSM0wHWZwvWRwcPgFjhyHyY5eN12hjfxl2ZtN0oq1AA==","signature_status":"signed_v1","signed_at":"2026-07-05T02:05:25.147201Z","signed_message":"canonical_sha256_bytes"},"source_id":"2004.14070","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ecb233964a531abc991592ad950e49d57740640e5aaf3f5fb50f9c1a93be0eba","sha256:f302fe518b5cd1b8ed33cce69f6184dd68ed7a58116a05ae030b11a8769c0b75"],"state_sha256":"7e728e835fefbe748532a51064e0137d23d7c6ec6df443bec6231376f8bd59d4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vXOKD/zE1qER/Z+y3IJIKFZA5C1UCj4ssYhIcQlNL2v3/czeE+MDpVjiCil//n+tKhLG0ikpiLq1ORLr+sxHDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T21:44:38.190382Z","bundle_sha256":"1e88a619e65b853c936a86e5ca40b1ae9284a29de18851dd582f82ffd30739fa"}}