{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:WTZZNVHSAMCWVYJZBUCEDW7HTX","short_pith_number":"pith:WTZZNVHS","schema_version":"1.0","canonical_sha256":"b4f396d4f203056ae1390d0441dbe79dc1925b260eb9a23c36cefe752ee4c8bc","source":{"kind":"arxiv","id":"1709.07979","version":3},"attestation_state":"computed","paper":{"title":"Multi-task Learning with Gradient Guided Policy Specialization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"C. Karen Liu, Greg Turk, Wenhao Yu","submitted_at":"2017-09-23T00:54:18Z","abstract_excerpt":"We present a method for efficient learning of control policies for multiple related robotic motor skills. Our approach consists of two stages, joint training and specialization training. During the joint training stage, a neural network policy is trained with minimal information to disambiguate the motor skills. This forces the policy to learn a common representation of the different tasks. Then, during the specialization training stage we selectively split the weights of the policy based on a per-weight metric that measures the disagreement among the multiple tasks. By splitting part of the c"},"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":"1709.07979","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2017-09-23T00:54:18Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"82fd0339975c3f3eb0aef4dbe12bbbb0673f7123a9dd281969d665b8867cf067","abstract_canon_sha256":"addc6ca8047e7963aaaaf0fb2605a54ac6b287d12e14a14da11b60ed413bf751"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:22:05.649819Z","signature_b64":"agU3sI7m9YM8qMRMmavPVgFUkRXY1Nws0df8uZcsC0zqcVmVGnzuergsOGu5kV2yVTuD2mEwd0nLJ9/QtiMOBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b4f396d4f203056ae1390d0441dbe79dc1925b260eb9a23c36cefe752ee4c8bc","last_reissued_at":"2026-05-18T00:22:05.649147Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:22:05.649147Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-task Learning with Gradient Guided Policy Specialization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"C. Karen Liu, Greg Turk, Wenhao Yu","submitted_at":"2017-09-23T00:54:18Z","abstract_excerpt":"We present a method for efficient learning of control policies for multiple related robotic motor skills. Our approach consists of two stages, joint training and specialization training. During the joint training stage, a neural network policy is trained with minimal information to disambiguate the motor skills. This forces the policy to learn a common representation of the different tasks. Then, during the specialization training stage we selectively split the weights of the policy based on a per-weight metric that measures the disagreement among the multiple tasks. By splitting part of the c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1709.07979","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":""},"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":"1709.07979","created_at":"2026-05-18T00:22:05.649244+00:00"},{"alias_kind":"arxiv_version","alias_value":"1709.07979v3","created_at":"2026-05-18T00:22:05.649244+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1709.07979","created_at":"2026-05-18T00:22:05.649244+00:00"},{"alias_kind":"pith_short_12","alias_value":"WTZZNVHSAMCW","created_at":"2026-05-18T12:31:53.515858+00:00"},{"alias_kind":"pith_short_16","alias_value":"WTZZNVHSAMCWVYJZ","created_at":"2026-05-18T12:31:53.515858+00:00"},{"alias_kind":"pith_short_8","alias_value":"WTZZNVHS","created_at":"2026-05-18T12:31:53.515858+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WTZZNVHSAMCWVYJZBUCEDW7HTX","json":"https://pith.science/pith/WTZZNVHSAMCWVYJZBUCEDW7HTX.json","graph_json":"https://pith.science/api/pith-number/WTZZNVHSAMCWVYJZBUCEDW7HTX/graph.json","events_json":"https://pith.science/api/pith-number/WTZZNVHSAMCWVYJZBUCEDW7HTX/events.json","paper":"https://pith.science/paper/WTZZNVHS"},"agent_actions":{"view_html":"https://pith.science/pith/WTZZNVHSAMCWVYJZBUCEDW7HTX","download_json":"https://pith.science/pith/WTZZNVHSAMCWVYJZBUCEDW7HTX.json","view_paper":"https://pith.science/paper/WTZZNVHS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1709.07979&json=true","fetch_graph":"https://pith.science/api/pith-number/WTZZNVHSAMCWVYJZBUCEDW7HTX/graph.json","fetch_events":"https://pith.science/api/pith-number/WTZZNVHSAMCWVYJZBUCEDW7HTX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WTZZNVHSAMCWVYJZBUCEDW7HTX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WTZZNVHSAMCWVYJZBUCEDW7HTX/action/storage_attestation","attest_author":"https://pith.science/pith/WTZZNVHSAMCWVYJZBUCEDW7HTX/action/author_attestation","sign_citation":"https://pith.science/pith/WTZZNVHSAMCWVYJZBUCEDW7HTX/action/citation_signature","submit_replication":"https://pith.science/pith/WTZZNVHSAMCWVYJZBUCEDW7HTX/action/replication_record"}},"created_at":"2026-05-18T00:22:05.649244+00:00","updated_at":"2026-05-18T00:22:05.649244+00:00"}