{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:M6ICILQ2TIKRVNJ5SCFD6NVYTQ","short_pith_number":"pith:M6ICILQ2","canonical_record":{"source":{"id":"2103.06473","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-03-11T05:39:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a359449c136ca104414e974ee79baba9aa8fc96e2b8f5cf1bec507ca277cbd73","abstract_canon_sha256":"ce9a5eede0f81a180f121090f0d97f74005fa9535ea2b68ca03962caa4df9443"},"schema_version":"1.0"},"canonical_sha256":"6790242e1a9a151ab53d908a3f36b89c1a8dde973a2c3b366008390780a4c173","source":{"kind":"arxiv","id":"2103.06473","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2103.06473","created_at":"2026-07-05T02:22:09Z"},{"alias_kind":"arxiv_version","alias_value":"2103.06473v1","created_at":"2026-07-05T02:22:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.06473","created_at":"2026-07-05T02:22:09Z"},{"alias_kind":"pith_short_12","alias_value":"M6ICILQ2TIKR","created_at":"2026-07-05T02:22:09Z"},{"alias_kind":"pith_short_16","alias_value":"M6ICILQ2TIKRVNJ5","created_at":"2026-07-05T02:22:09Z"},{"alias_kind":"pith_short_8","alias_value":"M6ICILQ2","created_at":"2026-07-05T02:22:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:M6ICILQ2TIKRVNJ5SCFD6NVYTQ","target":"record","payload":{"canonical_record":{"source":{"id":"2103.06473","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-03-11T05:39:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a359449c136ca104414e974ee79baba9aa8fc96e2b8f5cf1bec507ca277cbd73","abstract_canon_sha256":"ce9a5eede0f81a180f121090f0d97f74005fa9535ea2b68ca03962caa4df9443"},"schema_version":"1.0"},"canonical_sha256":"6790242e1a9a151ab53d908a3f36b89c1a8dde973a2c3b366008390780a4c173","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:22:09.379825Z","signature_b64":"8oltS+meVmI+gPD0hM2qwVV9qUMLsWB4Wa0eUJO53gVXLQSN2GJB9YIwLeiBKEToK81sbYwRk2QlyQhidqK4AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6790242e1a9a151ab53d908a3f36b89c1a8dde973a2c3b366008390780a4c173","last_reissued_at":"2026-07-05T02:22:09.379377Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:22:09.379377Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2103.06473","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":"2026-07-05T02:22:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"K5nTPdHTeY8xRgQj/0EekwEOvnG6DCLuhz08w5QMzsMOuMxl5r7f5OBpwJWXdXW0BBTmcOmyO13ffO8wEG5sBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T20:21:05.969898Z"},"content_sha256":"4825f32e2f679a5597b2ec9fef004b00795dbd71c035ce343082fc22030912d1","schema_version":"1.0","event_id":"sha256:4825f32e2f679a5597b2ec9fef004b00795dbd71c035ce343082fc22030912d1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:M6ICILQ2TIKRVNJ5SCFD6NVYTQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Multi-Task Federated Reinforcement Learning with Adversaries","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Aqeel Anwar, Arijit Raychowdhury","submitted_at":"2021-03-11T05:39:52Z","abstract_excerpt":"Reinforcement learning algorithms, just like any other Machine learning algorithm pose a serious threat from adversaries. The adversaries can manipulate the learning algorithm resulting in non-optimal policies. In this paper, we analyze the Multi-task Federated Reinforcement Learning algorithms, where multiple collaborative agents in various environments are trying to maximize the sum of discounted return, in the presence of adversarial agents. We argue that the common attack methods are not guaranteed to carry out a successful attack on Multi-task Federated Reinforcement Learning and propose "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.06473","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/2103.06473/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":"2026-07-05T02:22:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MTJYWiQSsYRGRKrqHdaGJc483byrtdmwbSS6OoDoajHv83h+d6zU3knQ8/VEUejJwSbSa7H/o/hZ3pM+lRglCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T20:21:05.970269Z"},"content_sha256":"8c6929c19e1f27e98fd77c6ec7951d8d59d3ec5c8ce2ccb5c0bebe5ce3d5e836","schema_version":"1.0","event_id":"sha256:8c6929c19e1f27e98fd77c6ec7951d8d59d3ec5c8ce2ccb5c0bebe5ce3d5e836"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/M6ICILQ2TIKRVNJ5SCFD6NVYTQ/bundle.json","state_url":"https://pith.science/pith/M6ICILQ2TIKRVNJ5SCFD6NVYTQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/M6ICILQ2TIKRVNJ5SCFD6NVYTQ/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":"2026-08-17T20:21:05Z","links":{"resolver":"https://pith.science/pith/M6ICILQ2TIKRVNJ5SCFD6NVYTQ","bundle":"https://pith.science/pith/M6ICILQ2TIKRVNJ5SCFD6NVYTQ/bundle.json","state":"https://pith.science/pith/M6ICILQ2TIKRVNJ5SCFD6NVYTQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/M6ICILQ2TIKRVNJ5SCFD6NVYTQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:M6ICILQ2TIKRVNJ5SCFD6NVYTQ","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":"ce9a5eede0f81a180f121090f0d97f74005fa9535ea2b68ca03962caa4df9443","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-03-11T05:39:52Z","title_canon_sha256":"a359449c136ca104414e974ee79baba9aa8fc96e2b8f5cf1bec507ca277cbd73"},"schema_version":"1.0","source":{"id":"2103.06473","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2103.06473","created_at":"2026-07-05T02:22:09Z"},{"alias_kind":"arxiv_version","alias_value":"2103.06473v1","created_at":"2026-07-05T02:22:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.06473","created_at":"2026-07-05T02:22:09Z"},{"alias_kind":"pith_short_12","alias_value":"M6ICILQ2TIKR","created_at":"2026-07-05T02:22:09Z"},{"alias_kind":"pith_short_16","alias_value":"M6ICILQ2TIKRVNJ5","created_at":"2026-07-05T02:22:09Z"},{"alias_kind":"pith_short_8","alias_value":"M6ICILQ2","created_at":"2026-07-05T02:22:09Z"}],"graph_snapshots":[{"event_id":"sha256:8c6929c19e1f27e98fd77c6ec7951d8d59d3ec5c8ce2ccb5c0bebe5ce3d5e836","target":"graph","created_at":"2026-07-05T02:22:09Z","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/2103.06473/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning algorithms, just like any other Machine learning algorithm pose a serious threat from adversaries. The adversaries can manipulate the learning algorithm resulting in non-optimal policies. In this paper, we analyze the Multi-task Federated Reinforcement Learning algorithms, where multiple collaborative agents in various environments are trying to maximize the sum of discounted return, in the presence of adversarial agents. We argue that the common attack methods are not guaranteed to carry out a successful attack on Multi-task Federated Reinforcement Learning and propose ","authors_text":"Aqeel Anwar, Arijit Raychowdhury","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-03-11T05:39:52Z","title":"Multi-Task Federated Reinforcement Learning with Adversaries"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.06473","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":"sha256:4825f32e2f679a5597b2ec9fef004b00795dbd71c035ce343082fc22030912d1","target":"record","created_at":"2026-07-05T02:22:09Z","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":"ce9a5eede0f81a180f121090f0d97f74005fa9535ea2b68ca03962caa4df9443","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-03-11T05:39:52Z","title_canon_sha256":"a359449c136ca104414e974ee79baba9aa8fc96e2b8f5cf1bec507ca277cbd73"},"schema_version":"1.0","source":{"id":"2103.06473","kind":"arxiv","version":1}},"canonical_sha256":"6790242e1a9a151ab53d908a3f36b89c1a8dde973a2c3b366008390780a4c173","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6790242e1a9a151ab53d908a3f36b89c1a8dde973a2c3b366008390780a4c173","first_computed_at":"2026-07-05T02:22:09.379377Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:22:09.379377Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8oltS+meVmI+gPD0hM2qwVV9qUMLsWB4Wa0eUJO53gVXLQSN2GJB9YIwLeiBKEToK81sbYwRk2QlyQhidqK4AA==","signature_status":"signed_v1","signed_at":"2026-07-05T02:22:09.379825Z","signed_message":"canonical_sha256_bytes"},"source_id":"2103.06473","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4825f32e2f679a5597b2ec9fef004b00795dbd71c035ce343082fc22030912d1","sha256:8c6929c19e1f27e98fd77c6ec7951d8d59d3ec5c8ce2ccb5c0bebe5ce3d5e836"],"state_sha256":"4b5422458f8815183c78442de99278d56d22e119c725066090bfec9a3f3fdf96"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"d9S48PloP8Jx6gZ0qYiKVDS8ODKzEUgY6YFaRV7H9LWnXgogbGar6CqPSi8CgmrDhtMfkYojDAW9gudE5BPADA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T20:21:05.973214Z","bundle_sha256":"e0f0105503ae1609200f451a3cde8a6d3f81dae01f62ccebde11d3a518dd0a4e"}}