{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:E6JC2KJG5KTH7CJK54SYP2LD2W","short_pith_number":"pith:E6JC2KJG","schema_version":"1.0","canonical_sha256":"27922d2926eaa67f892aef2587e963d583a9c3120da37938af7dfc60c4cc27f7","source":{"kind":"arxiv","id":"2101.07123","version":1},"attestation_state":"computed","paper":{"title":"Learning Successor States and Goal-Dependent Values: A Mathematical Viewpoint","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Corentin Tallec, L\\'eonard Blier, Yann Ollivier","submitted_at":"2021-01-18T15:33:26Z","abstract_excerpt":"In reinforcement learning, temporal difference-based algorithms can be sample-inefficient: for instance, with sparse rewards, no learning occurs until a reward is observed. This can be remedied by learning richer objects, such as a model of the environment, or successor states. Successor states model the expected future state occupancy from any given state for a given policy and are related to goal-dependent value functions, which learn how to reach arbitrary states. We formally derive the temporal difference algorithm for successor state and goal-dependent value function learning, either for "},"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":"2101.07123","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-01-18T15:33:26Z","cross_cats_sorted":[],"title_canon_sha256":"7be20f542c0e979622edbec2f3aff82cc329301a88cc520e9cc41374765dd932","abstract_canon_sha256":"99e1a6a9e6a83ddf40841e1be40b680ea8946457d0b0aced42af9a9115a6f367"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:07:40.509242Z","signature_b64":"XqwhXO+FpRdlFiPgvYBew9BN8kxpX/QecM7odm6OiuPmcAU78JAyAbEOINzfqSHmIYQlGbrws1Os+qzqudZZDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"27922d2926eaa67f892aef2587e963d583a9c3120da37938af7dfc60c4cc27f7","last_reissued_at":"2026-07-05T02:07:40.508818Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:07:40.508818Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Successor States and Goal-Dependent Values: A Mathematical Viewpoint","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Corentin Tallec, L\\'eonard Blier, Yann Ollivier","submitted_at":"2021-01-18T15:33:26Z","abstract_excerpt":"In reinforcement learning, temporal difference-based algorithms can be sample-inefficient: for instance, with sparse rewards, no learning occurs until a reward is observed. This can be remedied by learning richer objects, such as a model of the environment, or successor states. Successor states model the expected future state occupancy from any given state for a given policy and are related to goal-dependent value functions, which learn how to reach arbitrary states. We formally derive the temporal difference algorithm for successor state and goal-dependent value function learning, either for "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2101.07123","kind":"arxiv","version":1},"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/2101.07123/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":"2101.07123","created_at":"2026-07-05T02:07:40.508895+00:00"},{"alias_kind":"arxiv_version","alias_value":"2101.07123v1","created_at":"2026-07-05T02:07:40.508895+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2101.07123","created_at":"2026-07-05T02:07:40.508895+00:00"},{"alias_kind":"pith_short_12","alias_value":"E6JC2KJG5KTH","created_at":"2026-07-05T02:07:40.508895+00:00"},{"alias_kind":"pith_short_16","alias_value":"E6JC2KJG5KTH7CJK","created_at":"2026-07-05T02:07:40.508895+00:00"},{"alias_kind":"pith_short_8","alias_value":"E6JC2KJG","created_at":"2026-07-05T02:07:40.508895+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.29980","citing_title":"Exploration and Online Transfer with Behavioral Foundation Models","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29906","citing_title":"Plan, Don't Pose: Long Composite Motion Generation with Text-Aligned BFM","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15603","citing_title":"Offline Reinforcement Learning with Universal Horizon Models","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17017","citing_title":"When Dynamics Shift, Robust Task Inference Wins: Offline Imitation Learning with Behavior Foundation Models Revisited","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2506.08902","citing_title":"Intention-Conditioned Flow Occupancy Models","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2506.10137","citing_title":"Self-Predictive Representations for Combinatorial Generalization in Behavioral Cloning","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2603.20103","citing_title":"Spectral Alignment in Forward-Backward Representations via Temporal Abstraction","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13207","citing_title":"Switching Successor Measures for Hierarchical Zero-shot Reinforcement Learning","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01862","citing_title":"QHyer: Q-conditioned Hybrid Attention-mamba Transformer for Offline Goal-conditioned RL","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18064","citing_title":"Understanding Human Actions through the Lens of Executable Models","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17551","citing_title":"SVL: Goal-Conditioned Reinforcement Learning as Survival Learning","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E6JC2KJG5KTH7CJK54SYP2LD2W","json":"https://pith.science/pith/E6JC2KJG5KTH7CJK54SYP2LD2W.json","graph_json":"https://pith.science/api/pith-number/E6JC2KJG5KTH7CJK54SYP2LD2W/graph.json","events_json":"https://pith.science/api/pith-number/E6JC2KJG5KTH7CJK54SYP2LD2W/events.json","paper":"https://pith.science/paper/E6JC2KJG"},"agent_actions":{"view_html":"https://pith.science/pith/E6JC2KJG5KTH7CJK54SYP2LD2W","download_json":"https://pith.science/pith/E6JC2KJG5KTH7CJK54SYP2LD2W.json","view_paper":"https://pith.science/paper/E6JC2KJG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2101.07123&json=true","fetch_graph":"https://pith.science/api/pith-number/E6JC2KJG5KTH7CJK54SYP2LD2W/graph.json","fetch_events":"https://pith.science/api/pith-number/E6JC2KJG5KTH7CJK54SYP2LD2W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E6JC2KJG5KTH7CJK54SYP2LD2W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E6JC2KJG5KTH7CJK54SYP2LD2W/action/storage_attestation","attest_author":"https://pith.science/pith/E6JC2KJG5KTH7CJK54SYP2LD2W/action/author_attestation","sign_citation":"https://pith.science/pith/E6JC2KJG5KTH7CJK54SYP2LD2W/action/citation_signature","submit_replication":"https://pith.science/pith/E6JC2KJG5KTH7CJK54SYP2LD2W/action/replication_record"}},"created_at":"2026-07-05T02:07:40.508895+00:00","updated_at":"2026-07-05T02:07:40.508895+00:00"}