{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:RCVWN2C6DIXAWN4DAKJOWVCXTH","short_pith_number":"pith:RCVWN2C6","canonical_record":{"source":{"id":"2501.06258","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-09T10:21:19Z","cross_cats_sorted":[],"title_canon_sha256":"641bc8ce5da1cb84aaa3661a1e673821dede860f63447edb3e923c0a1380b6fd","abstract_canon_sha256":"a0d6debeb403c8282110486e65d8cb549cd1e0911d4f09c7aad19272952c83cc"},"schema_version":"1.0"},"canonical_sha256":"88ab66e85e1a2e0b37830292eb545799c47c06f3930ee61bc244d49624415672","source":{"kind":"arxiv","id":"2501.06258","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.06258","created_at":"2026-07-05T10:00:00Z"},{"alias_kind":"arxiv_version","alias_value":"2501.06258v1","created_at":"2026-07-05T10:00:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.06258","created_at":"2026-07-05T10:00:00Z"},{"alias_kind":"pith_short_12","alias_value":"RCVWN2C6DIXA","created_at":"2026-07-05T10:00:00Z"},{"alias_kind":"pith_short_16","alias_value":"RCVWN2C6DIXAWN4D","created_at":"2026-07-05T10:00:00Z"},{"alias_kind":"pith_short_8","alias_value":"RCVWN2C6","created_at":"2026-07-05T10:00:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:RCVWN2C6DIXAWN4DAKJOWVCXTH","target":"record","payload":{"canonical_record":{"source":{"id":"2501.06258","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-09T10:21:19Z","cross_cats_sorted":[],"title_canon_sha256":"641bc8ce5da1cb84aaa3661a1e673821dede860f63447edb3e923c0a1380b6fd","abstract_canon_sha256":"a0d6debeb403c8282110486e65d8cb549cd1e0911d4f09c7aad19272952c83cc"},"schema_version":"1.0"},"canonical_sha256":"88ab66e85e1a2e0b37830292eb545799c47c06f3930ee61bc244d49624415672","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:00:00.009147Z","signature_b64":"lOaeVP7uYE+fbux58exGwdIQXHA4SQe5UP+9et0XDXA7BZvCnYA4lzTiAJzS14P4ic3suee6hQ9wh6cti/aiBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"88ab66e85e1a2e0b37830292eb545799c47c06f3930ee61bc244d49624415672","last_reissued_at":"2026-07-05T10:00:00.008716Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:00:00.008716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.06258","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:00:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"axalGAJ2ZoOkcThVXkoF2rgrh1RKI610wqzhIuJTgBcM7hfSpaUTIPrdBMLslLDRIawRJ5wrBOYMe/U8hm2KAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T15:40:38.987109Z"},"content_sha256":"2f36cd40dcce352f903fdb5729e842329560865627f25bd4bbbbb291d4252654","schema_version":"1.0","event_id":"sha256:2f36cd40dcce352f903fdb5729e842329560865627f25bd4bbbbb291d4252654"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:RCVWN2C6DIXAWN4DAKJOWVCXTH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Contextual Bandit Optimization with Pre-Trained Neural Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Mikhail Terekhov","submitted_at":"2025-01-09T10:21:19Z","abstract_excerpt":"Bandit optimization is a difficult problem, especially if the reward model is high-dimensional. When rewards are modeled by neural networks, sublinear regret has only been shown under strong assumptions, usually when the network is extremely wide. In this thesis, we investigate how pre-training can help us in the regime of smaller models. We consider a stochastic contextual bandit with the rewards modeled by a multi-layer neural network. The last layer is a linear predictor, and the layers before it are a black box neural architecture, which we call a representation network. We model pre-train"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.06258","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/2501.06258/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:00:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/p12LAgZq14EFJ3ZtWOOq9Oj2+Dxy9XD6V5X2zCvaDQd32syfwYy5L+L1n5rLjmUhpxbqCr3WdYDnG6gzXNkAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T15:40:38.987642Z"},"content_sha256":"0f959010dca820309cdeed23a7275872779b01bc7e66f8a7ffdd204bc519a95a","schema_version":"1.0","event_id":"sha256:0f959010dca820309cdeed23a7275872779b01bc7e66f8a7ffdd204bc519a95a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RCVWN2C6DIXAWN4DAKJOWVCXTH/bundle.json","state_url":"https://pith.science/pith/RCVWN2C6DIXAWN4DAKJOWVCXTH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RCVWN2C6DIXAWN4DAKJOWVCXTH/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-11T15:40:38Z","links":{"resolver":"https://pith.science/pith/RCVWN2C6DIXAWN4DAKJOWVCXTH","bundle":"https://pith.science/pith/RCVWN2C6DIXAWN4DAKJOWVCXTH/bundle.json","state":"https://pith.science/pith/RCVWN2C6DIXAWN4DAKJOWVCXTH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RCVWN2C6DIXAWN4DAKJOWVCXTH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:RCVWN2C6DIXAWN4DAKJOWVCXTH","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"a0d6debeb403c8282110486e65d8cb549cd1e0911d4f09c7aad19272952c83cc","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-09T10:21:19Z","title_canon_sha256":"641bc8ce5da1cb84aaa3661a1e673821dede860f63447edb3e923c0a1380b6fd"},"schema_version":"1.0","source":{"id":"2501.06258","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.06258","created_at":"2026-07-05T10:00:00Z"},{"alias_kind":"arxiv_version","alias_value":"2501.06258v1","created_at":"2026-07-05T10:00:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.06258","created_at":"2026-07-05T10:00:00Z"},{"alias_kind":"pith_short_12","alias_value":"RCVWN2C6DIXA","created_at":"2026-07-05T10:00:00Z"},{"alias_kind":"pith_short_16","alias_value":"RCVWN2C6DIXAWN4D","created_at":"2026-07-05T10:00:00Z"},{"alias_kind":"pith_short_8","alias_value":"RCVWN2C6","created_at":"2026-07-05T10:00:00Z"}],"graph_snapshots":[{"event_id":"sha256:0f959010dca820309cdeed23a7275872779b01bc7e66f8a7ffdd204bc519a95a","target":"graph","created_at":"2026-07-05T10:00:00Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2501.06258/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Bandit optimization is a difficult problem, especially if the reward model is high-dimensional. When rewards are modeled by neural networks, sublinear regret has only been shown under strong assumptions, usually when the network is extremely wide. In this thesis, we investigate how pre-training can help us in the regime of smaller models. We consider a stochastic contextual bandit with the rewards modeled by a multi-layer neural network. The last layer is a linear predictor, and the layers before it are a black box neural architecture, which we call a representation network. We model pre-train","authors_text":"Mikhail Terekhov","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-09T10:21:19Z","title":"Contextual Bandit Optimization with Pre-Trained Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.06258","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:2f36cd40dcce352f903fdb5729e842329560865627f25bd4bbbbb291d4252654","target":"record","created_at":"2026-07-05T10:00:00Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"a0d6debeb403c8282110486e65d8cb549cd1e0911d4f09c7aad19272952c83cc","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-09T10:21:19Z","title_canon_sha256":"641bc8ce5da1cb84aaa3661a1e673821dede860f63447edb3e923c0a1380b6fd"},"schema_version":"1.0","source":{"id":"2501.06258","kind":"arxiv","version":1}},"canonical_sha256":"88ab66e85e1a2e0b37830292eb545799c47c06f3930ee61bc244d49624415672","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"88ab66e85e1a2e0b37830292eb545799c47c06f3930ee61bc244d49624415672","first_computed_at":"2026-07-05T10:00:00.008716Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:00:00.008716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lOaeVP7uYE+fbux58exGwdIQXHA4SQe5UP+9et0XDXA7BZvCnYA4lzTiAJzS14P4ic3suee6hQ9wh6cti/aiBA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:00:00.009147Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.06258","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2f36cd40dcce352f903fdb5729e842329560865627f25bd4bbbbb291d4252654","sha256:0f959010dca820309cdeed23a7275872779b01bc7e66f8a7ffdd204bc519a95a"],"state_sha256":"f8faf89d08f9eaabb5380f82264c85a24e5975ff7bad36d5ffbc24cd6a279f35"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0odZAubTxeI1xQliMCWhUhEchsXaRsppYb4AUBeGhvjRjlg8/qd+UbzGxyhdiNe+yn77XfT3e9BCqXAXRDZLAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T15:40:38.991946Z","bundle_sha256":"be5a34c543ae6be107fea2f7dca493d95d9b6a700ad74178715d8bf6c05ede9a"}}