{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:XCNXVJNBTBTF4ADJSMVCECAHAN","short_pith_number":"pith:XCNXVJNB","schema_version":"1.0","canonical_sha256":"b89b7aa5a198665e0069932a220807036b842460a61e124ec23416ac76dd53ce","source":{"kind":"arxiv","id":"2012.08621","version":1},"attestation_state":"computed","paper":{"title":"BeBold: Exploration Beyond the Boundary of Explored Regions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Huazhe Xu, Joseph E. Gonzalez, Kurt Keutzer, Tianjun Zhang, Xiaolong Wang, Yi Wu, Yuandong Tian","submitted_at":"2020-12-15T21:26:54Z","abstract_excerpt":"Efficient exploration under sparse rewards remains a key challenge in deep reinforcement learning. To guide exploration, previous work makes extensive use of intrinsic reward (IR). There are many heuristics for IR, including visitation counts, curiosity, and state-difference. In this paper, we analyze the pros and cons of each method and propose the regulated difference of inverse visitation counts as a simple but effective criterion for IR. The criterion helps the agent explore Beyond the Boundary of explored regions and mitigates common issues in count-based methods, such as short-sightednes"},"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":"2012.08621","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-12-15T21:26:54Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"1e1617a43febb3892a6098fc66e3f3f831a14c6be418ed763e4b7a6b456adb70","abstract_canon_sha256":"cc97e3872118f8b32f173b846f22f8aac4ed99ce976e4541772899a1da4ad1f5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:00:01.373218Z","signature_b64":"iJlESWume+u9TBBQmQxh/QD9YeZs9E7MAFBcB9Bp2/8u2c63RCN1PwDwg2jrUzYeKCsAIBH372x/xPn6ZFRXDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b89b7aa5a198665e0069932a220807036b842460a61e124ec23416ac76dd53ce","last_reissued_at":"2026-07-05T02:00:01.372756Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:00:01.372756Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BeBold: Exploration Beyond the Boundary of Explored Regions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Huazhe Xu, Joseph E. Gonzalez, Kurt Keutzer, Tianjun Zhang, Xiaolong Wang, Yi Wu, Yuandong Tian","submitted_at":"2020-12-15T21:26:54Z","abstract_excerpt":"Efficient exploration under sparse rewards remains a key challenge in deep reinforcement learning. To guide exploration, previous work makes extensive use of intrinsic reward (IR). There are many heuristics for IR, including visitation counts, curiosity, and state-difference. In this paper, we analyze the pros and cons of each method and propose the regulated difference of inverse visitation counts as a simple but effective criterion for IR. The criterion helps the agent explore Beyond the Boundary of explored regions and mitigates common issues in count-based methods, such as short-sightednes"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.08621","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/2012.08621/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":"2012.08621","created_at":"2026-07-05T02:00:01.372817+00:00"},{"alias_kind":"arxiv_version","alias_value":"2012.08621v1","created_at":"2026-07-05T02:00:01.372817+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.08621","created_at":"2026-07-05T02:00:01.372817+00:00"},{"alias_kind":"pith_short_12","alias_value":"XCNXVJNBTBTF","created_at":"2026-07-05T02:00:01.372817+00:00"},{"alias_kind":"pith_short_16","alias_value":"XCNXVJNBTBTF4ADJ","created_at":"2026-07-05T02:00:01.372817+00:00"},{"alias_kind":"pith_short_8","alias_value":"XCNXVJNB","created_at":"2026-07-05T02:00:01.372817+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.15414","citing_title":"Beyond Single-Model Optimization: Preserving Plasticity in Continual Reinforcement Learning","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XCNXVJNBTBTF4ADJSMVCECAHAN","json":"https://pith.science/pith/XCNXVJNBTBTF4ADJSMVCECAHAN.json","graph_json":"https://pith.science/api/pith-number/XCNXVJNBTBTF4ADJSMVCECAHAN/graph.json","events_json":"https://pith.science/api/pith-number/XCNXVJNBTBTF4ADJSMVCECAHAN/events.json","paper":"https://pith.science/paper/XCNXVJNB"},"agent_actions":{"view_html":"https://pith.science/pith/XCNXVJNBTBTF4ADJSMVCECAHAN","download_json":"https://pith.science/pith/XCNXVJNBTBTF4ADJSMVCECAHAN.json","view_paper":"https://pith.science/paper/XCNXVJNB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2012.08621&json=true","fetch_graph":"https://pith.science/api/pith-number/XCNXVJNBTBTF4ADJSMVCECAHAN/graph.json","fetch_events":"https://pith.science/api/pith-number/XCNXVJNBTBTF4ADJSMVCECAHAN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XCNXVJNBTBTF4ADJSMVCECAHAN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XCNXVJNBTBTF4ADJSMVCECAHAN/action/storage_attestation","attest_author":"https://pith.science/pith/XCNXVJNBTBTF4ADJSMVCECAHAN/action/author_attestation","sign_citation":"https://pith.science/pith/XCNXVJNBTBTF4ADJSMVCECAHAN/action/citation_signature","submit_replication":"https://pith.science/pith/XCNXVJNBTBTF4ADJSMVCECAHAN/action/replication_record"}},"created_at":"2026-07-05T02:00:01.372817+00:00","updated_at":"2026-07-05T02:00:01.372817+00:00"}