{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:7WWN55BARQG7BYH3CJMFWNXTXV","short_pith_number":"pith:7WWN55BA","schema_version":"1.0","canonical_sha256":"fdacdef4208c0df0e0fb12585b36f3bd7ad96fdac8fe07927a25989ec780a6ef","source":{"kind":"arxiv","id":"2106.14876","version":1},"attestation_state":"computed","paper":{"title":"Multi-task curriculum learning in a complex, visual, hard-exploration domain: Minecraft","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Adrien Ecoffet, Bowen Baker, Brandon Houghton, David Farhi, Ingmar Kanitscheider, Jeff Clune, Jie Tang, Joost Huizinga, Oleg Klimov, Peter Zhokhov, Raul Sampedro, William Hebgen Guss","submitted_at":"2021-06-28T17:50:40Z","abstract_excerpt":"An important challenge in reinforcement learning is training agents that can solve a wide variety of tasks. If tasks depend on each other (e.g. needing to learn to walk before learning to run), curriculum learning can speed up learning by focusing on the next best task to learn. We explore curriculum learning in a complex, visual domain with many hard exploration challenges: Minecraft. We find that learning progress (defined as a change in success probability of a task) is a reliable measure of learnability for automatically constructing an effective curriculum. We introduce a learning-progres"},"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":"2106.14876","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-28T17:50:40Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"548f379f17234583a822bafcb64c858ff80cc14d8f0336970d446715ac7d7bb0","abstract_canon_sha256":"8dada83bb2df7f60c08e0dde79b074be94aa462bd88953e13879cfc6176c3caf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:52:54.746689Z","signature_b64":"Uv5xU+XT9dm6YSGUpNlo2CZPn57WUHO4TTeBf4hdTas4jw3AsoQfGmEe+V47k/ChOwcdEWAyw3rXhenbvFbaBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fdacdef4208c0df0e0fb12585b36f3bd7ad96fdac8fe07927a25989ec780a6ef","last_reissued_at":"2026-07-05T02:52:54.746182Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:52:54.746182Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-task curriculum learning in a complex, visual, hard-exploration domain: Minecraft","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Adrien Ecoffet, Bowen Baker, Brandon Houghton, David Farhi, Ingmar Kanitscheider, Jeff Clune, Jie Tang, Joost Huizinga, Oleg Klimov, Peter Zhokhov, Raul Sampedro, William Hebgen Guss","submitted_at":"2021-06-28T17:50:40Z","abstract_excerpt":"An important challenge in reinforcement learning is training agents that can solve a wide variety of tasks. If tasks depend on each other (e.g. needing to learn to walk before learning to run), curriculum learning can speed up learning by focusing on the next best task to learn. We explore curriculum learning in a complex, visual domain with many hard exploration challenges: Minecraft. We find that learning progress (defined as a change in success probability of a task) is a reliable measure of learnability for automatically constructing an effective curriculum. We introduce a learning-progres"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.14876","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/2106.14876/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":"2106.14876","created_at":"2026-07-05T02:52:54.746241+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.14876v1","created_at":"2026-07-05T02:52:54.746241+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.14876","created_at":"2026-07-05T02:52:54.746241+00:00"},{"alias_kind":"pith_short_12","alias_value":"7WWN55BARQG7","created_at":"2026-07-05T02:52:54.746241+00:00"},{"alias_kind":"pith_short_16","alias_value":"7WWN55BARQG7BYH3","created_at":"2026-07-05T02:52:54.746241+00:00"},{"alias_kind":"pith_short_8","alias_value":"7WWN55BA","created_at":"2026-07-05T02:52:54.746241+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14831","citing_title":"Interestingness as an Inductive Heuristic for Future Compression Progress","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2305.17144","citing_title":"Ghost in the Minecraft: Generally Capable Agents for Open-World Environments via Large Language Models with Text-based Knowledge and Memory","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14350","citing_title":"Distributionally Robust Multi-Task Reinforcement Learning via Adaptive Task Sampling","ref_index":286,"is_internal_anchor":false},{"citing_arxiv_id":"2509.24527","citing_title":"Training Agents Inside of Scalable World Models","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2301.04104","citing_title":"Mastering Diverse Domains through World Models","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2305.16291","citing_title":"Voyager: An Open-Ended Embodied Agent with Large Language Models","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7WWN55BARQG7BYH3CJMFWNXTXV","json":"https://pith.science/pith/7WWN55BARQG7BYH3CJMFWNXTXV.json","graph_json":"https://pith.science/api/pith-number/7WWN55BARQG7BYH3CJMFWNXTXV/graph.json","events_json":"https://pith.science/api/pith-number/7WWN55BARQG7BYH3CJMFWNXTXV/events.json","paper":"https://pith.science/paper/7WWN55BA"},"agent_actions":{"view_html":"https://pith.science/pith/7WWN55BARQG7BYH3CJMFWNXTXV","download_json":"https://pith.science/pith/7WWN55BARQG7BYH3CJMFWNXTXV.json","view_paper":"https://pith.science/paper/7WWN55BA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.14876&json=true","fetch_graph":"https://pith.science/api/pith-number/7WWN55BARQG7BYH3CJMFWNXTXV/graph.json","fetch_events":"https://pith.science/api/pith-number/7WWN55BARQG7BYH3CJMFWNXTXV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7WWN55BARQG7BYH3CJMFWNXTXV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7WWN55BARQG7BYH3CJMFWNXTXV/action/storage_attestation","attest_author":"https://pith.science/pith/7WWN55BARQG7BYH3CJMFWNXTXV/action/author_attestation","sign_citation":"https://pith.science/pith/7WWN55BARQG7BYH3CJMFWNXTXV/action/citation_signature","submit_replication":"https://pith.science/pith/7WWN55BARQG7BYH3CJMFWNXTXV/action/replication_record"}},"created_at":"2026-07-05T02:52:54.746241+00:00","updated_at":"2026-07-05T02:52:54.746241+00:00"}