{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:Z2HWYZ3KUGXLXLHJGGZGEWZPPW","short_pith_number":"pith:Z2HWYZ3K","schema_version":"1.0","canonical_sha256":"ce8f6c676aa1aebbace931b2625b2f7db3b530b6728761d8f73835429ffbe0c4","source":{"kind":"arxiv","id":"1806.07377","version":6},"attestation_state":"computed","paper":{"title":"Transfer Learning for Related Reinforcement Learning Tasks via Image-to-Image Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Shani Gamrian, Yoav Goldberg","submitted_at":"2018-05-31T09:25:29Z","abstract_excerpt":"Despite the remarkable success of Deep RL in learning control policies from raw pixels, the resulting models do not generalize. We demonstrate that a trained agent fails completely when facing small visual changes, and that fine-tuning---the common transfer learning paradigm---fails to adapt to these changes, to the extent that it is faster to re-train the model from scratch. We show that by separating the visual transfer task from the control policy we achieve substantially better sample efficiency and transfer behavior, allowing an agent trained on the source task to transfer well to the tar"},"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":"1806.07377","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-05-31T09:25:29Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"51c1537473e3bddf50f4734f420172a593df6ea31ead890c0c620a30002b356e","abstract_canon_sha256":"256dd5e08e06eda131152c801e8f9c00a9f66b73ba7cc9a38878c6fefc85729e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:41:31.208619Z","signature_b64":"oJ6UfuOAEf/cYuAOw+pfOVdkV2v4PPqV2mhPJLYlHQmD8CZ+VeL7ilO6Z7TZGKzWixrU6RiULXNgujJy5LT8Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ce8f6c676aa1aebbace931b2625b2f7db3b530b6728761d8f73835429ffbe0c4","last_reissued_at":"2026-05-17T23:41:31.207894Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:41:31.207894Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Transfer Learning for Related Reinforcement Learning Tasks via Image-to-Image Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Shani Gamrian, Yoav Goldberg","submitted_at":"2018-05-31T09:25:29Z","abstract_excerpt":"Despite the remarkable success of Deep RL in learning control policies from raw pixels, the resulting models do not generalize. We demonstrate that a trained agent fails completely when facing small visual changes, and that fine-tuning---the common transfer learning paradigm---fails to adapt to these changes, to the extent that it is faster to re-train the model from scratch. We show that by separating the visual transfer task from the control policy we achieve substantially better sample efficiency and transfer behavior, allowing an agent trained on the source task to transfer well to the tar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.07377","kind":"arxiv","version":6},"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":"1806.07377","created_at":"2026-05-17T23:41:31.208006+00:00"},{"alias_kind":"arxiv_version","alias_value":"1806.07377v6","created_at":"2026-05-17T23:41:31.208006+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.07377","created_at":"2026-05-17T23:41:31.208006+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z2HWYZ3KUGXL","created_at":"2026-05-18T12:33:04.347982+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z2HWYZ3KUGXLXLHJ","created_at":"2026-05-18T12:33:04.347982+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z2HWYZ3K","created_at":"2026-05-18T12:33:04.347982+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1907.09319","citing_title":"VRLS: A Unified Reinforcement Learning Scheduler for Vehicle-to-Vehicle Communications","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z2HWYZ3KUGXLXLHJGGZGEWZPPW","json":"https://pith.science/pith/Z2HWYZ3KUGXLXLHJGGZGEWZPPW.json","graph_json":"https://pith.science/api/pith-number/Z2HWYZ3KUGXLXLHJGGZGEWZPPW/graph.json","events_json":"https://pith.science/api/pith-number/Z2HWYZ3KUGXLXLHJGGZGEWZPPW/events.json","paper":"https://pith.science/paper/Z2HWYZ3K"},"agent_actions":{"view_html":"https://pith.science/pith/Z2HWYZ3KUGXLXLHJGGZGEWZPPW","download_json":"https://pith.science/pith/Z2HWYZ3KUGXLXLHJGGZGEWZPPW.json","view_paper":"https://pith.science/paper/Z2HWYZ3K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1806.07377&json=true","fetch_graph":"https://pith.science/api/pith-number/Z2HWYZ3KUGXLXLHJGGZGEWZPPW/graph.json","fetch_events":"https://pith.science/api/pith-number/Z2HWYZ3KUGXLXLHJGGZGEWZPPW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z2HWYZ3KUGXLXLHJGGZGEWZPPW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z2HWYZ3KUGXLXLHJGGZGEWZPPW/action/storage_attestation","attest_author":"https://pith.science/pith/Z2HWYZ3KUGXLXLHJGGZGEWZPPW/action/author_attestation","sign_citation":"https://pith.science/pith/Z2HWYZ3KUGXLXLHJGGZGEWZPPW/action/citation_signature","submit_replication":"https://pith.science/pith/Z2HWYZ3KUGXLXLHJGGZGEWZPPW/action/replication_record"}},"created_at":"2026-05-17T23:41:31.208006+00:00","updated_at":"2026-05-17T23:41:31.208006+00:00"}