{"id":"https://openalex.org/W4308019808","doi":"https://doi.org/10.48550/arxiv.2211.00241","title":"Adversarial Policies Beat Superhuman Go AIs","display_name":"Adversarial Policies Beat Superhuman Go AIs","publication_year":2022,"publication_date":"2022-11-01","ids":{"openalex":"https://openalex.org/W4308019808","doi":"https://doi.org/10.48550/arxiv.2211.00241"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2211.00241","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2211.00241","pdf_url":"https://arxiv.org/pdf/2211.00241","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2211.00241","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5021249854","display_name":"Tony Tong Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Tony T.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067087948","display_name":"Adam Gleave","orcid":"https://orcid.org/0009-0001-6994-1103","observed_orcids":["https://orcid.org/0009-0001-6994-1103"]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gleave, Adam","raw_affiliation_strings":["FAR AI","UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"FAR AI","institution_ids":[]},{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087045282","display_name":"Tom Tseng","orcid":"https://orcid.org/0000-0002-6422-288X","observed_orcids":["https://orcid.org/0000-0002-6422-288X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tseng, Tom","raw_affiliation_strings":["FAR AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"FAR AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Pelrine, Kellin","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I5023651","display_name":"McGill University","ror":"https://ror.org/01pxwe438","country_code":"CA","type":"education","lineage":["https://openalex.org/I5023651"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Pelrine, Kellin","raw_affiliation_strings":["FAR AI","McGill University ; Mila"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"FAR AI","institution_ids":[]},{"raw_affiliation_string":"McGill University ; Mila","institution_ids":["https://openalex.org/I5023651"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088767744","display_name":"Nora Belrose","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Belrose, Nora","raw_affiliation_strings":["FAR AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"FAR AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008160642","display_name":"Joseph Michael Miller","orcid":"https://orcid.org/0000-0002-4545-0190","observed_orcids":["https://orcid.org/0000-0002-4545-0190","https://orcid.org/0009-0002-8836-8311"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Miller, Joseph","raw_affiliation_strings":["FAR AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"FAR AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Dennis, Michael D.","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dennis, Michael D.","raw_affiliation_strings":["UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102008535","display_name":"Yawen Duan","orcid":"https://orcid.org/0000-0002-5124-1192","observed_orcids":["https://orcid.org/0000-0002-5124-1192"]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Duan, Yawen","raw_affiliation_strings":["UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004361442","display_name":"Viktor Pogrebniak","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pogrebniak, Viktor","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026322200","display_name":"Sergey Levine","orcid":"https://orcid.org/0000-0001-6764-2743","observed_orcids":["https://orcid.org/0000-0001-6764-2743"]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Levine, Sergey","raw_affiliation_strings":["UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5104083277","display_name":"Stuart Russell","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Russell, Stuart","raw_affiliation_strings":["UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":54,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9955999851226807,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9955999851226807,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11574","display_name":"Artificial Intelligence in Games","score":0.0035000001080334187,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11424","display_name":"Security and Verification in Computing","score":0.00019999999494757503,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/ai-failure-modes","display_name":"AI failure modes","score":0.5989999771118164},{"id":"https://openalex.org/keywords/adversarial-training","display_name":"adversarial training","score":0.5429999828338623},{"id":"https://openalex.org/keywords/zero-shot-transfer","display_name":"zero-shot transfer","score":0.640999972820282}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.7689192891120911},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.5100956559181213},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.48690855503082275},{"id":"https://openalex.org/C189809214","wikidata":"https://www.wikidata.org/wiki/Q829522","display_name":"Beat (acoustics)","level":2,"score":0.4404948949813843},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.22995105385780334},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C24890656","wikidata":"https://www.wikidata.org/wiki/Q82811","display_name":"Acoustics","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2211.00241","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2211.00241","pdf_url":"https://arxiv.org/pdf/2211.00241","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2211.00241","doi":"https://doi.org/10.48550/arxiv.2211.00241","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2211.00241","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2211.00241","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2211.00241","pdf_url":"https://arxiv.org/pdf/2211.00241","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4308019808.pdf","grobid_xml":"https://content.openalex.org/works/W4308019808.grobid-xml"},"referenced_works_count":9,"referenced_works":["https://openalex.org/W2586751528","https://openalex.org/W2900994053","https://openalex.org/W2913848079","https://openalex.org/W2918318309","https://openalex.org/W2996037775","https://openalex.org/W3018951278","https://openalex.org/W4307930337","https://openalex.org/W4310705874","https://openalex.org/W4388711601"],"related_works":["https://openalex.org/W4409361542","https://openalex.org/W7133243399","https://openalex.org/W4409053427","https://openalex.org/W2945664122","https://openalex.org/W4391618815","https://openalex.org/W3091503989","https://openalex.org/W2257979135","https://openalex.org/W6992551938","https://openalex.org/W4307909381","https://openalex.org/W4400414497"],"abstract_inverted_index":{"We":[0],"attack":[1,43,77],"the":[2,55],"state-of-the-art":[3],"Go-playing":[4,49],"AI":[5,96],"system":[6],"KataGo":[7,20,37,81],"by":[8,30,75],"training":[9],"adversarial":[10],"policies":[11],"against":[12,19,87],"it,":[13],"achieving":[14],"a":[15],">97%":[16],"win":[17,29],"rate":[18],"running":[21],"at":[22],"superhuman":[23,48,69,95],"settings.":[24],"Our":[25,42,90],"adversaries":[26],"do":[27],"not":[28],"playing":[31],"Go":[32],"well.":[33],"Instead,":[34],"they":[35],"trick":[36],"into":[38],"making":[39],"serious":[40],"blunders.":[41],"transfers":[44],"zero-shot":[45],"to":[46,54,66,85],"other":[47],"AIs,":[50],"and":[51],"is":[52],"comprehensible":[53],"extent":[56],"that":[57,93],"human":[58],"experts":[59],"can":[60],"implement":[61],"it":[62],"without":[63],"algorithmic":[64],"assistance":[65],"consistently":[67],"beat":[68],"AIs.":[70],"The":[71],"core":[72],"vulnerability":[73],"uncovered":[74],"our":[76,88],"persists":[78],"even":[79,94],"in":[80],"agents":[82],"adversarially":[83],"trained":[84],"defend":[86],"attack.":[89],"results":[91],"demonstrate":[92],"systems":[97],"may":[98],"harbor":[99],"surprising":[100],"failure":[101],"modes.":[102],"Example":[103],"games":[104],"are":[105],"available":[106],"https://goattack.far.ai/.":[107]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":9},{"year":2024,"cited_by_count":19},{"year":2023,"cited_by_count":20},{"year":2022,"cited_by_count":2}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2022-11-07T00:00:00"}