{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:W4HCSEYKTTF7SDRWPUIAHI5MRP","short_pith_number":"pith:W4HCSEYK","schema_version":"1.0","canonical_sha256":"b70e29130a9ccbf90e367d1003a3ac8bcfe5542db47e60b7cbf7b26d2a108595","source":{"kind":"arxiv","id":"1811.01146","version":3},"attestation_state":"computed","paper":{"title":"Closed-Loop Memory GAN for Continual Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Amanda Rios, Laurent Itti","submitted_at":"2018-11-03T02:40:01Z","abstract_excerpt":"Sequential learning of tasks using gradient descent leads to an unremitting decline in the accuracy of tasks for which training data is no longer available, termed catastrophic forgetting. Generative models have been explored as a means to approximate the distribution of old tasks and bypass storage of real data. Here we propose a cumulative closed-loop memory replay GAN (CloGAN) provided with external regularization by a small memory unit selected for maximum sample diversity. We evaluate incremental class learning using a notoriously hard paradigm, single-headed learning, in which each task "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1811.01146","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-03T02:40:01Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"d79fd35659e35d1798f0a34c329963e2dae97745071039cde955554c30491e59","abstract_canon_sha256":"454b62d6efe5d392b3ca1b6177bc3383936f30e8d6f1a461dd7230daa0451b13"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:38:32.481428Z","signature_b64":"E45O0+cWMDPiSKunMPlPgjc5j7Bkp12GWDQsYDzV/PT9IYMX4YVrScMYwEsgVWUw2o7Qc0bx2x/8JopchDkEAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b70e29130a9ccbf90e367d1003a3ac8bcfe5542db47e60b7cbf7b26d2a108595","last_reissued_at":"2026-07-05T01:38:32.481000Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:38:32.481000Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Closed-Loop Memory GAN for Continual Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Amanda Rios, Laurent Itti","submitted_at":"2018-11-03T02:40:01Z","abstract_excerpt":"Sequential learning of tasks using gradient descent leads to an unremitting decline in the accuracy of tasks for which training data is no longer available, termed catastrophic forgetting. Generative models have been explored as a means to approximate the distribution of old tasks and bypass storage of real data. Here we propose a cumulative closed-loop memory replay GAN (CloGAN) provided with external regularization by a small memory unit selected for maximum sample diversity. We evaluate incremental class learning using a notoriously hard paradigm, single-headed learning, in which each task "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.01146","kind":"arxiv","version":3},"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/1811.01146/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"1811.01146","created_at":"2026-07-05T01:38:32.481051+00:00"},{"alias_kind":"arxiv_version","alias_value":"1811.01146v3","created_at":"2026-07-05T01:38:32.481051+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.01146","created_at":"2026-07-05T01:38:32.481051+00:00"},{"alias_kind":"pith_short_12","alias_value":"W4HCSEYKTTF7","created_at":"2026-07-05T01:38:32.481051+00:00"},{"alias_kind":"pith_short_16","alias_value":"W4HCSEYKTTF7SDRW","created_at":"2026-07-05T01:38:32.481051+00:00"},{"alias_kind":"pith_short_8","alias_value":"W4HCSEYK","created_at":"2026-07-05T01:38:32.481051+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"1906.10528","citing_title":"Beneficial perturbation network for continual learning","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/W4HCSEYKTTF7SDRWPUIAHI5MRP","json":"https://pith.science/pith/W4HCSEYKTTF7SDRWPUIAHI5MRP.json","graph_json":"https://pith.science/api/pith-number/W4HCSEYKTTF7SDRWPUIAHI5MRP/graph.json","events_json":"https://pith.science/api/pith-number/W4HCSEYKTTF7SDRWPUIAHI5MRP/events.json","paper":"https://pith.science/paper/W4HCSEYK"},"agent_actions":{"view_html":"https://pith.science/pith/W4HCSEYKTTF7SDRWPUIAHI5MRP","download_json":"https://pith.science/pith/W4HCSEYKTTF7SDRWPUIAHI5MRP.json","view_paper":"https://pith.science/paper/W4HCSEYK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1811.01146&json=true","fetch_graph":"https://pith.science/api/pith-number/W4HCSEYKTTF7SDRWPUIAHI5MRP/graph.json","fetch_events":"https://pith.science/api/pith-number/W4HCSEYKTTF7SDRWPUIAHI5MRP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/W4HCSEYKTTF7SDRWPUIAHI5MRP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/W4HCSEYKTTF7SDRWPUIAHI5MRP/action/storage_attestation","attest_author":"https://pith.science/pith/W4HCSEYKTTF7SDRWPUIAHI5MRP/action/author_attestation","sign_citation":"https://pith.science/pith/W4HCSEYKTTF7SDRWPUIAHI5MRP/action/citation_signature","submit_replication":"https://pith.science/pith/W4HCSEYKTTF7SDRWPUIAHI5MRP/action/replication_record"}},"created_at":"2026-07-05T01:38:32.481051+00:00","updated_at":"2026-07-05T01:38:32.481051+00:00"}