{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:EBUGCSATFPXEFHHAEFAPHXXZUC","short_pith_number":"pith:EBUGCSAT","schema_version":"1.0","canonical_sha256":"20686148132bee429ce02140f3def9a0ba33c5ad0bce7a8a5a6416bd21369b23","source":{"kind":"arxiv","id":"1810.02274","version":5},"attestation_state":"computed","paper":{"title":"Episodic Curiosity through Reachability","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.RO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Anton Raichuk, Damien Vincent, Marc Pollefeys, Nikolay Savinov, Rapha\\\"el Marinier, Sylvain Gelly, Timothy Lillicrap","submitted_at":"2018-10-04T15:24:06Z","abstract_excerpt":"Rewards are sparse in the real world and most of today's reinforcement learning algorithms struggle with such sparsity. One solution to this problem is to allow the agent to create rewards for itself - thus making rewards dense and more suitable for learning. In particular, inspired by curious behaviour in animals, observing something novel could be rewarded with a bonus. Such bonus is summed up with the real task reward - making it possible for RL algorithms to learn from the combined reward. We propose a new curiosity method which uses episodic memory to form the novelty bonus. To determine "},"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":"1810.02274","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-10-04T15:24:06Z","cross_cats_sorted":["cs.AI","cs.CV","cs.RO","stat.ML"],"title_canon_sha256":"236355d26a00e734d12fb04bdf2b2b2e83b8c700ba48520f78b2a040ec27b525","abstract_canon_sha256":"13b79f333ecb48928129845521442af90654161b7e0e6618fe06e455839b7595"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:51:36.269974Z","signature_b64":"SPURIcddJoOTxhohf3176A9FCOGYIVaU50UETkHyHWLPPvvK9BylYdGvZdugUOHRF7HFoa+3wTRtsJZesOvNDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"20686148132bee429ce02140f3def9a0ba33c5ad0bce7a8a5a6416bd21369b23","last_reissued_at":"2026-07-04T23:51:36.269550Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:51:36.269550Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Episodic Curiosity through Reachability","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.RO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Anton Raichuk, Damien Vincent, Marc Pollefeys, Nikolay Savinov, Rapha\\\"el Marinier, Sylvain Gelly, Timothy Lillicrap","submitted_at":"2018-10-04T15:24:06Z","abstract_excerpt":"Rewards are sparse in the real world and most of today's reinforcement learning algorithms struggle with such sparsity. One solution to this problem is to allow the agent to create rewards for itself - thus making rewards dense and more suitable for learning. In particular, inspired by curious behaviour in animals, observing something novel could be rewarded with a bonus. Such bonus is summed up with the real task reward - making it possible for RL algorithms to learn from the combined reward. We propose a new curiosity method which uses episodic memory to form the novelty bonus. To determine "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.02274","kind":"arxiv","version":5},"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/1810.02274/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":"1810.02274","created_at":"2026-07-04T23:51:36.269607+00:00"},{"alias_kind":"arxiv_version","alias_value":"1810.02274v5","created_at":"2026-07-04T23:51:36.269607+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.02274","created_at":"2026-07-04T23:51:36.269607+00:00"},{"alias_kind":"pith_short_12","alias_value":"EBUGCSATFPXE","created_at":"2026-07-04T23:51:36.269607+00:00"},{"alias_kind":"pith_short_16","alias_value":"EBUGCSATFPXEFHHA","created_at":"2026-07-04T23:51:36.269607+00:00"},{"alias_kind":"pith_short_8","alias_value":"EBUGCSAT","created_at":"2026-07-04T23:51:36.269607+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"1907.01285","citing_title":"Learning the Arrow of Time","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2509.25438","citing_title":"Beyond Noisy-TVs: Noise-Robust Exploration Via Learning Progress Monitoring","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EBUGCSATFPXEFHHAEFAPHXXZUC","json":"https://pith.science/pith/EBUGCSATFPXEFHHAEFAPHXXZUC.json","graph_json":"https://pith.science/api/pith-number/EBUGCSATFPXEFHHAEFAPHXXZUC/graph.json","events_json":"https://pith.science/api/pith-number/EBUGCSATFPXEFHHAEFAPHXXZUC/events.json","paper":"https://pith.science/paper/EBUGCSAT"},"agent_actions":{"view_html":"https://pith.science/pith/EBUGCSATFPXEFHHAEFAPHXXZUC","download_json":"https://pith.science/pith/EBUGCSATFPXEFHHAEFAPHXXZUC.json","view_paper":"https://pith.science/paper/EBUGCSAT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1810.02274&json=true","fetch_graph":"https://pith.science/api/pith-number/EBUGCSATFPXEFHHAEFAPHXXZUC/graph.json","fetch_events":"https://pith.science/api/pith-number/EBUGCSATFPXEFHHAEFAPHXXZUC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EBUGCSATFPXEFHHAEFAPHXXZUC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EBUGCSATFPXEFHHAEFAPHXXZUC/action/storage_attestation","attest_author":"https://pith.science/pith/EBUGCSATFPXEFHHAEFAPHXXZUC/action/author_attestation","sign_citation":"https://pith.science/pith/EBUGCSATFPXEFHHAEFAPHXXZUC/action/citation_signature","submit_replication":"https://pith.science/pith/EBUGCSATFPXEFHHAEFAPHXXZUC/action/replication_record"}},"created_at":"2026-07-04T23:51:36.269607+00:00","updated_at":"2026-07-04T23:51:36.269607+00:00"}