{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:MTEHG53OR4HB6YEH564BZ4JUCY","short_pith_number":"pith:MTEHG53O","schema_version":"1.0","canonical_sha256":"64c873776e8f0e1f6087efb81cf134161226d4f77efaa0753d85accb1162dc7a","source":{"kind":"arxiv","id":"2406.17718","version":1},"attestation_state":"computed","paper":{"title":"When does Self-Prediction help? Understanding Auxiliary Tasks in Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Amir-massoud Farahmand, Claas Voelcker, Igor Gilitschenski, Tyler Kastner","submitted_at":"2024-06-25T17:06:57Z","abstract_excerpt":"We investigate the impact of auxiliary learning tasks such as observation reconstruction and latent self-prediction on the representation learning problem in reinforcement learning. We also study how they interact with distractions and observation functions in the MDP. We provide a theoretical analysis of the learning dynamics of observation reconstruction, latent self-prediction, and TD learning in the presence of distractions and observation functions under linear model assumptions. With this formalization, we are able to explain why latent-self prediction is a helpful \\emph{auxiliary 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":"2406.17718","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-06-25T17:06:57Z","cross_cats_sorted":[],"title_canon_sha256":"70ea05de14122cfd37e2a2a323d5df6e4c10ccdee432893cef376f66e649fc8a","abstract_canon_sha256":"cb17fe49d5b4829faccc8ae074935e3a2470e51aed3eff77d604da9e6e98635e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:36:39.094240Z","signature_b64":"hg79cm5MIH4ZKSOgkZw8o+DDYzQRe0D5ljUate1TChmgS6bzgDsrxI5cJo3pNlOYsbvQLFurs7CNWpf9NdiqBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"64c873776e8f0e1f6087efb81cf134161226d4f77efaa0753d85accb1162dc7a","last_reissued_at":"2026-07-05T08:36:39.093833Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:36:39.093833Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When does Self-Prediction help? Understanding Auxiliary Tasks in Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Amir-massoud Farahmand, Claas Voelcker, Igor Gilitschenski, Tyler Kastner","submitted_at":"2024-06-25T17:06:57Z","abstract_excerpt":"We investigate the impact of auxiliary learning tasks such as observation reconstruction and latent self-prediction on the representation learning problem in reinforcement learning. We also study how they interact with distractions and observation functions in the MDP. We provide a theoretical analysis of the learning dynamics of observation reconstruction, latent self-prediction, and TD learning in the presence of distractions and observation functions under linear model assumptions. With this formalization, we are able to explain why latent-self prediction is a helpful \\emph{auxiliary task},"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.17718","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/2406.17718/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":"2406.17718","created_at":"2026-07-05T08:36:39.093887+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.17718v1","created_at":"2026-07-05T08:36:39.093887+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.17718","created_at":"2026-07-05T08:36:39.093887+00:00"},{"alias_kind":"pith_short_12","alias_value":"MTEHG53OR4HB","created_at":"2026-07-05T08:36:39.093887+00:00"},{"alias_kind":"pith_short_16","alias_value":"MTEHG53OR4HB6YEH","created_at":"2026-07-05T08:36:39.093887+00:00"},{"alias_kind":"pith_short_8","alias_value":"MTEHG53O","created_at":"2026-07-05T08:36:39.093887+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.10137","citing_title":"Self-Predictive Representations for Combinatorial Generalization in Behavioral Cloning","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2603.07083","citing_title":"Dreamer-CDP: Improving Reconstruction-free World Models Via Continuous Deterministic Representation Prediction","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03023","citing_title":"Behavior-Constrained Reinforcement Learning with Receding-Horizon Credit Assignment for High-Performance Control","ref_index":44,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MTEHG53OR4HB6YEH564BZ4JUCY","json":"https://pith.science/pith/MTEHG53OR4HB6YEH564BZ4JUCY.json","graph_json":"https://pith.science/api/pith-number/MTEHG53OR4HB6YEH564BZ4JUCY/graph.json","events_json":"https://pith.science/api/pith-number/MTEHG53OR4HB6YEH564BZ4JUCY/events.json","paper":"https://pith.science/paper/MTEHG53O"},"agent_actions":{"view_html":"https://pith.science/pith/MTEHG53OR4HB6YEH564BZ4JUCY","download_json":"https://pith.science/pith/MTEHG53OR4HB6YEH564BZ4JUCY.json","view_paper":"https://pith.science/paper/MTEHG53O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.17718&json=true","fetch_graph":"https://pith.science/api/pith-number/MTEHG53OR4HB6YEH564BZ4JUCY/graph.json","fetch_events":"https://pith.science/api/pith-number/MTEHG53OR4HB6YEH564BZ4JUCY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MTEHG53OR4HB6YEH564BZ4JUCY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MTEHG53OR4HB6YEH564BZ4JUCY/action/storage_attestation","attest_author":"https://pith.science/pith/MTEHG53OR4HB6YEH564BZ4JUCY/action/author_attestation","sign_citation":"https://pith.science/pith/MTEHG53OR4HB6YEH564BZ4JUCY/action/citation_signature","submit_replication":"https://pith.science/pith/MTEHG53OR4HB6YEH564BZ4JUCY/action/replication_record"}},"created_at":"2026-07-05T08:36:39.093887+00:00","updated_at":"2026-07-05T08:36:39.093887+00:00"}