{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:AOVNQZGY6JYMVERPBZLMUAIALK","short_pith_number":"pith:AOVNQZGY","schema_version":"1.0","canonical_sha256":"03aad864d8f270ca922f0e56ca01005aac4c7368fd390a451e205820781a9518","source":{"kind":"arxiv","id":"2010.01069","version":4},"attestation_state":"computed","paper":{"title":"A Deeper Look at Discounting Mismatch in Actor-Critic Algorithms","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Harm van Seijen, Remi Tachet des Combes, Romain Laroche, Shangtong Zhang, Shimon Whiteson","submitted_at":"2020-10-02T15:51:48Z","abstract_excerpt":"We investigate the discounting mismatch in actor-critic algorithm implementations from a representation learning perspective. Theoretically, actor-critic algorithms usually have discounting for both actor and critic, i.e., there is a $\\gamma^t$ term in the actor update for the transition observed at time $t$ in a trajectory and the critic is a discounted value function. Practitioners, however, usually ignore the discounting ($\\gamma^t$) for the actor while using a discounted critic. We investigate this mismatch in two scenarios. In the first scenario, we consider optimizing an undiscounted obj"},"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":"2010.01069","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-10-02T15:51:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6a72d0d1bb0ba7484b98acd5efdd75e4e16cafef6a3a97956475c8c521931169","abstract_canon_sha256":"8e54e3fe64a25439d9440c995a3cdf7b32f3b16497e4db3d10994464d6885ed5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:51:41.066930Z","signature_b64":"wqOJhTae+OmJ36coS/Ylxm1f/AGQLrKWdLyhTXEZ2i/ASNQp9aZa3LsXe/0zjiy33otQfG5JJHy70304G0iICg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"03aad864d8f270ca922f0e56ca01005aac4c7368fd390a451e205820781a9518","last_reissued_at":"2026-07-05T03:51:41.066462Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:51:41.066462Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Deeper Look at Discounting Mismatch in Actor-Critic Algorithms","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Harm van Seijen, Remi Tachet des Combes, Romain Laroche, Shangtong Zhang, Shimon Whiteson","submitted_at":"2020-10-02T15:51:48Z","abstract_excerpt":"We investigate the discounting mismatch in actor-critic algorithm implementations from a representation learning perspective. Theoretically, actor-critic algorithms usually have discounting for both actor and critic, i.e., there is a $\\gamma^t$ term in the actor update for the transition observed at time $t$ in a trajectory and the critic is a discounted value function. Practitioners, however, usually ignore the discounting ($\\gamma^t$) for the actor while using a discounted critic. We investigate this mismatch in two scenarios. In the first scenario, we consider optimizing an undiscounted obj"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.01069","kind":"arxiv","version":4},"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/2010.01069/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":"2010.01069","created_at":"2026-07-05T03:51:41.066525+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.01069v4","created_at":"2026-07-05T03:51:41.066525+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.01069","created_at":"2026-07-05T03:51:41.066525+00:00"},{"alias_kind":"pith_short_12","alias_value":"AOVNQZGY6JYM","created_at":"2026-07-05T03:51:41.066525+00:00"},{"alias_kind":"pith_short_16","alias_value":"AOVNQZGY6JYMVERP","created_at":"2026-07-05T03:51:41.066525+00:00"},{"alias_kind":"pith_short_8","alias_value":"AOVNQZGY","created_at":"2026-07-05T03:51:41.066525+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.06700","citing_title":"Average Reward Reinforcement Learning for Wireless Radio Resource Management","ref_index":30,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AOVNQZGY6JYMVERPBZLMUAIALK","json":"https://pith.science/pith/AOVNQZGY6JYMVERPBZLMUAIALK.json","graph_json":"https://pith.science/api/pith-number/AOVNQZGY6JYMVERPBZLMUAIALK/graph.json","events_json":"https://pith.science/api/pith-number/AOVNQZGY6JYMVERPBZLMUAIALK/events.json","paper":"https://pith.science/paper/AOVNQZGY"},"agent_actions":{"view_html":"https://pith.science/pith/AOVNQZGY6JYMVERPBZLMUAIALK","download_json":"https://pith.science/pith/AOVNQZGY6JYMVERPBZLMUAIALK.json","view_paper":"https://pith.science/paper/AOVNQZGY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.01069&json=true","fetch_graph":"https://pith.science/api/pith-number/AOVNQZGY6JYMVERPBZLMUAIALK/graph.json","fetch_events":"https://pith.science/api/pith-number/AOVNQZGY6JYMVERPBZLMUAIALK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AOVNQZGY6JYMVERPBZLMUAIALK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AOVNQZGY6JYMVERPBZLMUAIALK/action/storage_attestation","attest_author":"https://pith.science/pith/AOVNQZGY6JYMVERPBZLMUAIALK/action/author_attestation","sign_citation":"https://pith.science/pith/AOVNQZGY6JYMVERPBZLMUAIALK/action/citation_signature","submit_replication":"https://pith.science/pith/AOVNQZGY6JYMVERPBZLMUAIALK/action/replication_record"}},"created_at":"2026-07-05T03:51:41.066525+00:00","updated_at":"2026-07-05T03:51:41.066525+00:00"}