{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:G5ABKPQWSFIILB55WQSX5D36CZ","short_pith_number":"pith:G5ABKPQW","schema_version":"1.0","canonical_sha256":"3740153e1691508587bdb4257e8f7e16563a56f44a265098bdb0d5838d3238d2","source":{"kind":"arxiv","id":"2208.10481","version":1},"attestation_state":"computed","paper":{"title":"BARReL: Bottleneck Attention for Adversarial Robustness in Vision-Based Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR","cs.CV","cs.RO"],"primary_cat":"cs.LG","authors_text":"Daniel A. Keim, Eugene Bykovets, Joachim M. Buhmann, Mennatallah El-Assady, Yannick Metz","submitted_at":"2022-08-22T17:54:34Z","abstract_excerpt":"Robustness to adversarial perturbations has been explored in many areas of computer vision. This robustness is particularly relevant in vision-based reinforcement learning, as the actions of autonomous agents might be safety-critic or impactful in the real world. We investigate the susceptibility of vision-based reinforcement learning agents to gradient-based adversarial attacks and evaluate a potential defense. We observe that Bottleneck Attention Modules (BAM) included in CNN architectures can act as potential tools to increase robustness against adversarial attacks. We show how learned atte"},"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":"2208.10481","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-08-22T17:54:34Z","cross_cats_sorted":["cs.AI","cs.CR","cs.CV","cs.RO"],"title_canon_sha256":"74e24127fbfea6a8d4add7165ab0d705ab234a1f50e2f5355d22c37c2a0c006c","abstract_canon_sha256":"6912a230ea5b8a05174f41c55de67cd0b69a6df8b2677e7a35fb4cf7518cd9c1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:50:29.264430Z","signature_b64":"u7Qbncqt0gd1NUyPtosymmFo5ix2ROg5XhbtXNzOs3Owx2BmKOso6CvSs/mnjuoU4DDPfg0pzZ1OFLxFpElhAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3740153e1691508587bdb4257e8f7e16563a56f44a265098bdb0d5838d3238d2","last_reissued_at":"2026-07-05T04:50:29.264093Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:50:29.264093Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BARReL: Bottleneck Attention for Adversarial Robustness in Vision-Based Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR","cs.CV","cs.RO"],"primary_cat":"cs.LG","authors_text":"Daniel A. Keim, Eugene Bykovets, Joachim M. Buhmann, Mennatallah El-Assady, Yannick Metz","submitted_at":"2022-08-22T17:54:34Z","abstract_excerpt":"Robustness to adversarial perturbations has been explored in many areas of computer vision. This robustness is particularly relevant in vision-based reinforcement learning, as the actions of autonomous agents might be safety-critic or impactful in the real world. We investigate the susceptibility of vision-based reinforcement learning agents to gradient-based adversarial attacks and evaluate a potential defense. We observe that Bottleneck Attention Modules (BAM) included in CNN architectures can act as potential tools to increase robustness against adversarial attacks. We show how learned atte"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2208.10481","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/2208.10481/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":"2208.10481","created_at":"2026-07-05T04:50:29.264146+00:00"},{"alias_kind":"arxiv_version","alias_value":"2208.10481v1","created_at":"2026-07-05T04:50:29.264146+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2208.10481","created_at":"2026-07-05T04:50:29.264146+00:00"},{"alias_kind":"pith_short_12","alias_value":"G5ABKPQWSFII","created_at":"2026-07-05T04:50:29.264146+00:00"},{"alias_kind":"pith_short_16","alias_value":"G5ABKPQWSFIILB55","created_at":"2026-07-05T04:50:29.264146+00:00"},{"alias_kind":"pith_short_8","alias_value":"G5ABKPQW","created_at":"2026-07-05T04:50:29.264146+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/G5ABKPQWSFIILB55WQSX5D36CZ","json":"https://pith.science/pith/G5ABKPQWSFIILB55WQSX5D36CZ.json","graph_json":"https://pith.science/api/pith-number/G5ABKPQWSFIILB55WQSX5D36CZ/graph.json","events_json":"https://pith.science/api/pith-number/G5ABKPQWSFIILB55WQSX5D36CZ/events.json","paper":"https://pith.science/paper/G5ABKPQW"},"agent_actions":{"view_html":"https://pith.science/pith/G5ABKPQWSFIILB55WQSX5D36CZ","download_json":"https://pith.science/pith/G5ABKPQWSFIILB55WQSX5D36CZ.json","view_paper":"https://pith.science/paper/G5ABKPQW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2208.10481&json=true","fetch_graph":"https://pith.science/api/pith-number/G5ABKPQWSFIILB55WQSX5D36CZ/graph.json","fetch_events":"https://pith.science/api/pith-number/G5ABKPQWSFIILB55WQSX5D36CZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G5ABKPQWSFIILB55WQSX5D36CZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G5ABKPQWSFIILB55WQSX5D36CZ/action/storage_attestation","attest_author":"https://pith.science/pith/G5ABKPQWSFIILB55WQSX5D36CZ/action/author_attestation","sign_citation":"https://pith.science/pith/G5ABKPQWSFIILB55WQSX5D36CZ/action/citation_signature","submit_replication":"https://pith.science/pith/G5ABKPQWSFIILB55WQSX5D36CZ/action/replication_record"}},"created_at":"2026-07-05T04:50:29.264146+00:00","updated_at":"2026-07-05T04:50:29.264146+00:00"}