{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:QMVPKD55OKVAG3D2K3CYZI3Q5X","short_pith_number":"pith:QMVPKD55","schema_version":"1.0","canonical_sha256":"832af50fbd72aa036c7a56c58ca370edee5ef5f9a257f5d770f57e800be00724","source":{"kind":"arxiv","id":"1711.05769","version":2},"attestation_state":"computed","paper":{"title":"PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Arun Mallya, Svetlana Lazebnik","submitted_at":"2017-11-15T19:36:51Z","abstract_excerpt":"This paper presents a method for adding multiple tasks to a single deep neural network while avoiding catastrophic forgetting. Inspired by network pruning techniques, we exploit redundancies in large deep networks to free up parameters that can then be employed to learn new tasks. By performing iterative pruning and network re-training, we are able to sequentially \"pack\" multiple tasks into a single network while ensuring minimal drop in performance and minimal storage overhead. Unlike prior work that uses proxy losses to maintain accuracy on older tasks, we always optimize for the task at han"},"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":"1711.05769","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-11-15T19:36:51Z","cross_cats_sorted":[],"title_canon_sha256":"ca21c7beca1f147a00f921e56c6f4adb59103c26fec292c992d8126bc0e4bd10","abstract_canon_sha256":"5abdccc5ab3e9586e7c41b9d85ed4e9dff46977a724fbb8ebfef4ed181e5c2dc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:16:08.331347Z","signature_b64":"X+a42NUUlyu4gr3kW0vuL+0bTROXOvCffdnt132ZQls6bPLobm2U7vY1e8pxu4OzhH2sqWLjhQKbrcgSrd2SDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"832af50fbd72aa036c7a56c58ca370edee5ef5f9a257f5d770f57e800be00724","last_reissued_at":"2026-05-18T00:16:08.330755Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:16:08.330755Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Arun Mallya, Svetlana Lazebnik","submitted_at":"2017-11-15T19:36:51Z","abstract_excerpt":"This paper presents a method for adding multiple tasks to a single deep neural network while avoiding catastrophic forgetting. Inspired by network pruning techniques, we exploit redundancies in large deep networks to free up parameters that can then be employed to learn new tasks. By performing iterative pruning and network re-training, we are able to sequentially \"pack\" multiple tasks into a single network while ensuring minimal drop in performance and minimal storage overhead. Unlike prior work that uses proxy losses to maintain accuracy on older tasks, we always optimize for the task at han"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1711.05769","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"},"aliases":[{"alias_kind":"arxiv","alias_value":"1711.05769","created_at":"2026-05-18T00:16:08.330844+00:00"},{"alias_kind":"arxiv_version","alias_value":"1711.05769v2","created_at":"2026-05-18T00:16:08.330844+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1711.05769","created_at":"2026-05-18T00:16:08.330844+00:00"},{"alias_kind":"pith_short_12","alias_value":"QMVPKD55OKVA","created_at":"2026-05-18T12:31:39.905425+00:00"},{"alias_kind":"pith_short_16","alias_value":"QMVPKD55OKVAG3D2","created_at":"2026-05-18T12:31:39.905425+00:00"},{"alias_kind":"pith_short_8","alias_value":"QMVPKD55","created_at":"2026-05-18T12:31:39.905425+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2606.27374","citing_title":"World Action Models Enable Continual Imitation Learning with Recurrent Generative Replays","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2508.18187","citing_title":"BRAIN: Bias-Mitigation Continual Learning Approach to Vision-Brain Understanding","ref_index":114,"is_internal_anchor":true},{"citing_arxiv_id":"2605.01542","citing_title":"Mesh Based Simulations with Spatial and Temporal awareness","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QMVPKD55OKVAG3D2K3CYZI3Q5X","json":"https://pith.science/pith/QMVPKD55OKVAG3D2K3CYZI3Q5X.json","graph_json":"https://pith.science/api/pith-number/QMVPKD55OKVAG3D2K3CYZI3Q5X/graph.json","events_json":"https://pith.science/api/pith-number/QMVPKD55OKVAG3D2K3CYZI3Q5X/events.json","paper":"https://pith.science/paper/QMVPKD55"},"agent_actions":{"view_html":"https://pith.science/pith/QMVPKD55OKVAG3D2K3CYZI3Q5X","download_json":"https://pith.science/pith/QMVPKD55OKVAG3D2K3CYZI3Q5X.json","view_paper":"https://pith.science/paper/QMVPKD55","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1711.05769&json=true","fetch_graph":"https://pith.science/api/pith-number/QMVPKD55OKVAG3D2K3CYZI3Q5X/graph.json","fetch_events":"https://pith.science/api/pith-number/QMVPKD55OKVAG3D2K3CYZI3Q5X/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QMVPKD55OKVAG3D2K3CYZI3Q5X/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QMVPKD55OKVAG3D2K3CYZI3Q5X/action/storage_attestation","attest_author":"https://pith.science/pith/QMVPKD55OKVAG3D2K3CYZI3Q5X/action/author_attestation","sign_citation":"https://pith.science/pith/QMVPKD55OKVAG3D2K3CYZI3Q5X/action/citation_signature","submit_replication":"https://pith.science/pith/QMVPKD55OKVAG3D2K3CYZI3Q5X/action/replication_record"}},"created_at":"2026-05-18T00:16:08.330844+00:00","updated_at":"2026-05-18T00:16:08.330844+00:00"}