{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OUCVG3NVPU5EWB7I2NCYEFRJG6","short_pith_number":"pith:OUCVG3NV","schema_version":"1.0","canonical_sha256":"7505536db57d3a4b07e8d34582162937a521d9c33e8cad3e992a578d269bf295","source":{"kind":"arxiv","id":"2511.07046","version":4},"attestation_state":"computed","paper":{"title":"Learning Quantized Continuous Controllers for Integer Hardware","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Christoph H. Lampert, Fabian Kresse","submitted_at":"2025-11-10T12:39:14Z","abstract_excerpt":"Deploying continuous-control reinforcement learning policies on embedded hardware requires meeting tight latency and power budgets. Small FPGAs can deliver these, but only if costly floating-point pipelines are avoided. We study quantization-aware training (QAT) of policies for integer inference and we present a learning-to-hardware pipeline that automatically selects low-bit policies and synthesizes them to an Artix-7 FPGA. Across five MuJoCo tasks, we obtain policy networks that are competitive with full precision (FP32) policies but require as few as 3 or even only 2 bits per weight, and pe"},"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":"2511.07046","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-11-10T12:39:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"020b05353aff0d943951834dfe4bdf7bdedc2b037ad55c803b7305b283147621","abstract_canon_sha256":"c9bbdad64a0bd5f9c7ed03bc0cdac389cf7aa93edcbc5a753fe9f769efed2a18"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T02:07:12.685895Z","signature_b64":"e7fx4dBLc62Q6bKA+tR2Pby5J7njybH9E3MX2Y4JvnWq32v4nU7TMgaXQSdKxOWTnt9ESecI4UYde4SdTlFUAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7505536db57d3a4b07e8d34582162937a521d9c33e8cad3e992a578d269bf295","last_reissued_at":"2026-06-09T02:07:12.684985Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T02:07:12.684985Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Quantized Continuous Controllers for Integer Hardware","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Christoph H. Lampert, Fabian Kresse","submitted_at":"2025-11-10T12:39:14Z","abstract_excerpt":"Deploying continuous-control reinforcement learning policies on embedded hardware requires meeting tight latency and power budgets. Small FPGAs can deliver these, but only if costly floating-point pipelines are avoided. We study quantization-aware training (QAT) of policies for integer inference and we present a learning-to-hardware pipeline that automatically selects low-bit policies and synthesizes them to an Artix-7 FPGA. Across five MuJoCo tasks, we obtain policy networks that are competitive with full precision (FP32) policies but require as few as 3 or even only 2 bits per weight, and pe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.07046","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/2511.07046/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":"2511.07046","created_at":"2026-06-09T02:07:12.685102+00:00"},{"alias_kind":"arxiv_version","alias_value":"2511.07046v4","created_at":"2026-06-09T02:07:12.685102+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.07046","created_at":"2026-06-09T02:07:12.685102+00:00"},{"alias_kind":"pith_short_12","alias_value":"OUCVG3NVPU5E","created_at":"2026-06-09T02:07:12.685102+00:00"},{"alias_kind":"pith_short_16","alias_value":"OUCVG3NVPU5EWB7I","created_at":"2026-06-09T02:07:12.685102+00:00"},{"alias_kind":"pith_short_8","alias_value":"OUCVG3NV","created_at":"2026-06-09T02:07:12.685102+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OUCVG3NVPU5EWB7I2NCYEFRJG6","json":"https://pith.science/pith/OUCVG3NVPU5EWB7I2NCYEFRJG6.json","graph_json":"https://pith.science/api/pith-number/OUCVG3NVPU5EWB7I2NCYEFRJG6/graph.json","events_json":"https://pith.science/api/pith-number/OUCVG3NVPU5EWB7I2NCYEFRJG6/events.json","paper":"https://pith.science/paper/OUCVG3NV"},"agent_actions":{"view_html":"https://pith.science/pith/OUCVG3NVPU5EWB7I2NCYEFRJG6","download_json":"https://pith.science/pith/OUCVG3NVPU5EWB7I2NCYEFRJG6.json","view_paper":"https://pith.science/paper/OUCVG3NV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2511.07046&json=true","fetch_graph":"https://pith.science/api/pith-number/OUCVG3NVPU5EWB7I2NCYEFRJG6/graph.json","fetch_events":"https://pith.science/api/pith-number/OUCVG3NVPU5EWB7I2NCYEFRJG6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OUCVG3NVPU5EWB7I2NCYEFRJG6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OUCVG3NVPU5EWB7I2NCYEFRJG6/action/storage_attestation","attest_author":"https://pith.science/pith/OUCVG3NVPU5EWB7I2NCYEFRJG6/action/author_attestation","sign_citation":"https://pith.science/pith/OUCVG3NVPU5EWB7I2NCYEFRJG6/action/citation_signature","submit_replication":"https://pith.science/pith/OUCVG3NVPU5EWB7I2NCYEFRJG6/action/replication_record"}},"created_at":"2026-06-09T02:07:12.685102+00:00","updated_at":"2026-06-09T02:07:12.685102+00:00"}