{"id":"https://openalex.org/W4414359350","doi":"https://doi.org/10.24963/ijcai.2025/552","title":"Maximum Entropy Softmax Policy Gradient via Entropy Advantage Estimation","display_name":"Maximum Entropy Softmax Policy Gradient via Entropy Advantage Estimation","publication_year":2025,"publication_date":"2025-09-01","ids":{"openalex":"https://openalex.org/W4414359350","doi":"https://doi.org/10.24963/ijcai.2025/552"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2025/552","is_oa":false,"landing_page_url":"https://doi.org/10.24963/ijcai.2025/552","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5025797799","display_name":"Jean Seong Bjorn Choe","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jean Seong Bjorn Choe","raw_affiliation_strings":["Korea University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5068766413","display_name":"Jong-kook Kim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jong-kook Kim","raw_affiliation_strings":["Korea University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.11844848,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"4958","last_page":"4966"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.8166999816894531,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.8166999816894531,"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/T10067","display_name":"Stochastic processes and financial applications","score":0.7911999821662903,"subfield":{"id":"https://openalex.org/subfields/2003","display_name":"Finance"},"field":{"id":"https://openalex.org/fields/20","display_name":"Economics, Econometrics and Finance"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.7839000225067139,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/principle-of-maximum-entropy","display_name":"Principle of maximum entropy","score":0.7465000152587891},{"id":"https://openalex.org/keywords/softmax-function","display_name":"Softmax function","score":0.6498000025749207},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.605400025844574},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.4702000021934509},{"id":"https://openalex.org/keywords/maximum-entropy-spectral-estimation","display_name":"Maximum entropy spectral estimation","score":0.4115999937057495},{"id":"https://openalex.org/keywords/entropy-estimation","display_name":"Entropy estimation","score":0.3312999904155731}],"concepts":[{"id":"https://openalex.org/C9679016","wikidata":"https://www.wikidata.org/wiki/Q1417473","display_name":"Principle of maximum entropy","level":2,"score":0.7465000152587891},{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.6498000025749207},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.605400025844574},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5475999712944031},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.5232999920845032},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.4702000021934509},{"id":"https://openalex.org/C89735579","wikidata":"https://www.wikidata.org/wiki/Q6795894","display_name":"Maximum entropy spectral estimation","level":3,"score":0.4115999937057495},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.38339999318122864},{"id":"https://openalex.org/C95546049","wikidata":"https://www.wikidata.org/wiki/Q1345207","display_name":"Entropy estimation","level":3,"score":0.3312999904155731},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3149999976158142},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3095000088214874},{"id":"https://openalex.org/C196956702","wikidata":"https://www.wikidata.org/wiki/Q6795829","display_name":"Maximum-entropy Markov model","level":5,"score":0.2946000099182129},{"id":"https://openalex.org/C75782508","wikidata":"https://www.wikidata.org/wiki/Q3333633","display_name":"Cross-entropy method","level":4,"score":0.29170000553131104},{"id":"https://openalex.org/C167928553","wikidata":"https://www.wikidata.org/wiki/Q1376021","display_name":"Estimation theory","level":2,"score":0.29100000858306885},{"id":"https://openalex.org/C115680565","wikidata":"https://www.wikidata.org/wiki/Q5977448","display_name":"Gradient method","level":2,"score":0.27320000529289246},{"id":"https://openalex.org/C60507348","wikidata":"https://www.wikidata.org/wiki/Q6795892","display_name":"Maximum entropy probability distribution","level":3,"score":0.26649999618530273},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2644999921321869},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.263700008392334},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.2614000141620636}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.24963/ijcai.2025/552","is_oa":false,"landing_page_url":"https://doi.org/10.24963/ijcai.2025/552","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Entropy":[0,69],"Regularisation":[1],"is":[2,54],"a":[3,74,122],"widely":[4],"adopted":[5],"technique":[6],"that":[7,52,77,93],"enhances":[8],"policy":[9,21,35,44],"optimisation":[10,110],"performance":[11],"and":[12,86,99,113,124],"stability.":[13],"Maximum":[14],"entropy":[15,29,62,87],"reinforcement":[16],"learning":[17],"(MaxEnt":[18],"RL)":[19],"regularises":[20],"evaluation":[22],"by":[23,82],"augmenting":[24],"the":[25,57,61,106],"objective":[26],"with":[27],"an":[28],"term,":[30],"showing":[31],"theoretical":[32],"benefits":[33],"in":[34,41,64,115],"optimisation.":[36],"However,":[37],"its":[38],"practical":[39],"application":[40],"straightforward":[42],"direct":[43],"gradient":[45],"settings":[46],"remains":[47],"surprisingly":[48],"underexplored.":[49],"We":[50],"hypothesise":[51],"this":[53],"due":[55],"to":[56],"difficulty":[58],"of":[59],"managing":[60],"reward":[63],"practice.":[65],"This":[66],"paper":[67],"proposes":[68],"Advantage":[70],"Policy":[71,96,102],"Optimisation":[72,97,103],"(EAPO),":[73],"simple":[75],"method":[76,120],"facilitates":[78],"MaxEnt":[79,107,126],"RL":[80,127],"implementation":[81],"separately":[83],"estimating":[84],"task":[85],"objectives.":[88],"Our":[89],"empirical":[90],"evaluations":[91],"demonstrate":[92],"extending":[94],"Proximal":[95],"(PPO)":[98],"Trust":[100],"Region":[101],"(TRPO)":[104],"within":[105],"framework":[108],"improves":[109],"performance,":[111],"generalisation,":[112],"exploration":[114],"various":[116],"environments.":[117],"Moreover,":[118],"our":[119],"provides":[121],"stable":[123],"performant":[125],"algorithm":[128],"for":[129],"discrete":[130],"action":[131],"spaces.":[132]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}