{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VJZ2IFDTQRV5V5KEIAXDQDTNQD","short_pith_number":"pith:VJZ2IFDT","schema_version":"1.0","canonical_sha256":"aa73a41473846bdaf544402e380e6d80e4a01a609d0c99fcd7d91fe9933aacd9","source":{"kind":"arxiv","id":"2412.17113","version":1},"attestation_state":"computed","paper":{"title":"Adam on Local Time: Addressing Nonstationarity in RL with Relative Adam Timesteps","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Alexander D. Goldie, Andrei Lupu, Benjamin Ellis, Jakob Foerster, Matthew T. Jackson, Mattie Fellows, Shimon Whiteson","submitted_at":"2024-12-22T18:01:08Z","abstract_excerpt":"In reinforcement learning (RL), it is common to apply techniques used broadly in machine learning such as neural network function approximators and momentum-based optimizers. However, such tools were largely developed for supervised learning rather than nonstationary RL, leading practitioners to adopt target networks, clipped policy updates, and other RL-specific implementation tricks to combat this mismatch, rather than directly adapting this toolchain for use in RL. In this paper, we take a different approach and instead address the effect of nonstationarity by adapting the widely used Adam "},"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":"2412.17113","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-12-22T18:01:08Z","cross_cats_sorted":["math.OC"],"title_canon_sha256":"094156a9bc9f2a052d900da5d6298ce929efc748d8d0133936cad25f1c9a131e","abstract_canon_sha256":"5179179103ce2ca0fa84f08c5c2a6db193c0bd54f74da7de1737c9e1ab72d336"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:53:17.743848Z","signature_b64":"7IhffH+hS8Ld3k//dkYbYWQNJJg0H+dt9gmAfEnjCP+1xKQk6AGHkdsxmlIb166Oe9NLb+sx7jZDJnIQ48gHBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aa73a41473846bdaf544402e380e6d80e4a01a609d0c99fcd7d91fe9933aacd9","last_reissued_at":"2026-07-05T09:53:17.743371Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:53:17.743371Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adam on Local Time: Addressing Nonstationarity in RL with Relative Adam Timesteps","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Alexander D. Goldie, Andrei Lupu, Benjamin Ellis, Jakob Foerster, Matthew T. Jackson, Mattie Fellows, Shimon Whiteson","submitted_at":"2024-12-22T18:01:08Z","abstract_excerpt":"In reinforcement learning (RL), it is common to apply techniques used broadly in machine learning such as neural network function approximators and momentum-based optimizers. However, such tools were largely developed for supervised learning rather than nonstationary RL, leading practitioners to adopt target networks, clipped policy updates, and other RL-specific implementation tricks to combat this mismatch, rather than directly adapting this toolchain for use in RL. In this paper, we take a different approach and instead address the effect of nonstationarity by adapting the widely used Adam "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.17113","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/2412.17113/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":"2412.17113","created_at":"2026-07-05T09:53:17.743422+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.17113v1","created_at":"2026-07-05T09:53:17.743422+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.17113","created_at":"2026-07-05T09:53:17.743422+00:00"},{"alias_kind":"pith_short_12","alias_value":"VJZ2IFDTQRV5","created_at":"2026-07-05T09:53:17.743422+00:00"},{"alias_kind":"pith_short_16","alias_value":"VJZ2IFDTQRV5V5KE","created_at":"2026-07-05T09:53:17.743422+00:00"},{"alias_kind":"pith_short_8","alias_value":"VJZ2IFDT","created_at":"2026-07-05T09:53:17.743422+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.17668","citing_title":"How Should We Meta-Learn Reinforcement Learning Algorithms?","ref_index":19,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VJZ2IFDTQRV5V5KEIAXDQDTNQD","json":"https://pith.science/pith/VJZ2IFDTQRV5V5KEIAXDQDTNQD.json","graph_json":"https://pith.science/api/pith-number/VJZ2IFDTQRV5V5KEIAXDQDTNQD/graph.json","events_json":"https://pith.science/api/pith-number/VJZ2IFDTQRV5V5KEIAXDQDTNQD/events.json","paper":"https://pith.science/paper/VJZ2IFDT"},"agent_actions":{"view_html":"https://pith.science/pith/VJZ2IFDTQRV5V5KEIAXDQDTNQD","download_json":"https://pith.science/pith/VJZ2IFDTQRV5V5KEIAXDQDTNQD.json","view_paper":"https://pith.science/paper/VJZ2IFDT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.17113&json=true","fetch_graph":"https://pith.science/api/pith-number/VJZ2IFDTQRV5V5KEIAXDQDTNQD/graph.json","fetch_events":"https://pith.science/api/pith-number/VJZ2IFDTQRV5V5KEIAXDQDTNQD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VJZ2IFDTQRV5V5KEIAXDQDTNQD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VJZ2IFDTQRV5V5KEIAXDQDTNQD/action/storage_attestation","attest_author":"https://pith.science/pith/VJZ2IFDTQRV5V5KEIAXDQDTNQD/action/author_attestation","sign_citation":"https://pith.science/pith/VJZ2IFDTQRV5V5KEIAXDQDTNQD/action/citation_signature","submit_replication":"https://pith.science/pith/VJZ2IFDTQRV5V5KEIAXDQDTNQD/action/replication_record"}},"created_at":"2026-07-05T09:53:17.743422+00:00","updated_at":"2026-07-05T09:53:17.743422+00:00"}