{"id":"https://openalex.org/W2995894173","doi":null,"title":"Implementation Matters in Deep RL: A Case Study on PPO and TRPO","display_name":"Implementation Matters in Deep RL: A Case Study on PPO and TRPO","publication_year":2020,"publication_date":"2020-04-30","ids":{"openalex":"https://openalex.org/W2995894173","mag":"2995894173"},"language":"en","primary_location":{"id":"mag:2995894173","doi":null,"is_oa":false,"landing_page_url":"https://openreview.net/pdf?id=r1etN1rtPB","pdf_url":null,"source":{"id":"https://openalex.org/S4306419637","display_name":"International Conference on Learning Representations","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":"International Conference on Learning Representations","raw_type":null},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":[],"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/A5061993024","display_name":"Logan Engstrom","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Logan Engstrom","raw_affiliation_strings":["Massachusetts Institute of Technology, Cambridge, United States"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology, Cambridge, United States","institution_ids":["https://openalex.org/I63966007"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067427666","display_name":"Andrew Ilyas","orcid":"https://orcid.org/0000-0001-7187-5240","observed_orcids":["https://orcid.org/0000-0001-7187-5240"]},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Andrew Ilyas","raw_affiliation_strings":["Massachusetts Institute of Technology, Cambridge, United States"],"raw_orcid":"https://orcid.org/0000-0001-7187-5240","affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology, Cambridge, United States","institution_ids":["https://openalex.org/I63966007"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068204319","display_name":"Shibani Santurkar","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shibani Santurkar","raw_affiliation_strings":["Massachusetts Institute of Technology, Cambridge, United States"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology, Cambridge, United States","institution_ids":["https://openalex.org/I63966007"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031570233","display_name":"Dimitris Tsipras","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dimitris Tsipras","raw_affiliation_strings":["Massachusetts Institute of Technology, Cambridge, United States"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology, Cambridge, United States","institution_ids":["https://openalex.org/I63966007"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045352499","display_name":"Firdaus Janoos","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Firdaus Janoos","raw_affiliation_strings":["Massachusetts Institute of Technology, Cambridge, United States"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology, Cambridge, United States","institution_ids":["https://openalex.org/I63966007"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112072124","display_name":"Larry Rudolph","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Larry Rudolph","raw_affiliation_strings":["Massachusetts Institute of Technology, Cambridge, United States"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology, Cambridge, United States","institution_ids":["https://openalex.org/I63966007"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5032574693","display_name":"Aleksander M\u0105dry","orcid":"https://orcid.org/0000-0003-0536-0323","observed_orcids":["https://orcid.org/0000-0003-0536-0323"]},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Aleksander M\u0105dry","raw_affiliation_strings":["Massachusetts Institute of Technology, Cambridge, United States"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology, Cambridge, United States","institution_ids":["https://openalex.org/I63966007"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I63966007"],"apc_list":null,"apc_paid":null,"fwci":8.2379,"has_fulltext":false,"cited_by_count":154,"citation_normalized_percentile":{"value":0.9823309,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"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/T10462","display_name":"Reinforcement Learning in Robotics","score":1.0,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":1.0,"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":0.0,"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/T12794","display_name":"Adaptive Dynamic Programming Control","score":0.0,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/proximal-policy-optimization","display_name":"proximal policy optimization","score":0.9179999828338623},{"id":"https://openalex.org/keywords/trust-region-policy-optimization","display_name":"trust region policy optimization","score":0.9990000128746033},{"id":"https://openalex.org/keywords/performance-attribution","display_name":"performance attribution","score":0.6570000052452087},{"id":"https://openalex.org/keywords/cumulative-reward","display_name":"cumulative reward","score":0.7860000133514404},{"id":"https://openalex.org/keywords/agent-behavior","display_name":"agent behavior","score":0.5230000019073486}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.8859351277351379},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7573384642601013},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.7555277943611145},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.623268723487854},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.5022902488708496},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.46821191906929016},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46725034713745117},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.42741236090660095},{"id":"https://openalex.org/C89109886","wikidata":"https://www.wikidata.org/wiki/Q1535924","display_name":"Trust region","level":3,"score":0.41450393199920654},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.326241672039032},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.32326868176460266},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.13414448499679565},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11385592818260193},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.08523845672607422},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.06773972511291504},{"id":"https://openalex.org/C78458016","wikidata":"https://www.wikidata.org/wiki/Q840400","display_name":"Evolutionary biology","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C178635117","wikidata":"https://www.wikidata.org/wiki/Q747499","display_name":"RADIUS","level":2,"score":0.0},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"mag:2995894173","doi":null,"is_oa":false,"landing_page_url":"https://openreview.net/pdf?id=r1etN1rtPB","pdf_url":null,"source":{"id":"https://openalex.org/S4306419637","display_name":"International Conference on Learning Representations","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":"International Conference on Learning Representations","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":14,"referenced_works":["https://openalex.org/W1499669280","https://openalex.org/W1757796397","https://openalex.org/W1771410628","https://openalex.org/W2119717200","https://openalex.org/W2130801532","https://openalex.org/W2155027007","https://openalex.org/W2736601468","https://openalex.org/W2754517384","https://openalex.org/W2766447205","https://openalex.org/W2788837839","https://openalex.org/W2789525339","https://openalex.org/W2895212759","https://openalex.org/W2962749646","https://openalex.org/W2964121744"],"related_works":["https://openalex.org/W3032398409","https://openalex.org/W7048047797","https://openalex.org/W3143499011","https://openalex.org/W7095008446","https://openalex.org/W2965438240","https://openalex.org/W4391245004","https://openalex.org/W4409753347","https://openalex.org/W2970851609","https://openalex.org/W3167672220","https://openalex.org/W4411551875"],"abstract_inverted_index":{"We":[0,28],"study":[1,15],"the":[2,30,47,92],"roots":[3],"of":[4,32,51,73,96],"algorithmic":[5],"progress":[6],"in":[7,39,76,100],"deep":[8,101],"policy":[9],"gradient":[10],"algorithms":[11],"through":[12],"a":[13,57],"case":[14],"on":[16,60],"two":[17],"popular":[18],"algorithms,":[19],"Proximal":[20],"Policy":[21,26],"Optimization":[22],"and":[23,81,94],"Trust":[24],"Region":[25],"Optimization.":[27],"investigate":[29],"consequences":[31],"code-level":[33],"optimizations:":[34],"algorithm":[35],"augmentations":[36],"found":[37],"only":[38],"implementations":[40],"or":[41],"described":[42],"as":[43],"auxiliary":[44],"details":[45],"to":[46],"core":[48],"algorithm.":[49],"Seemingly":[50],"secondary":[52],"importance,":[53,95],"such":[54],"optimizations":[55],"have":[56],"major":[58],"impact":[59],"agent":[61],"behavior.":[62],"Our":[63],"results":[64],"show":[65,91],"that":[66],"they":[67],"(a)":[68],"are":[69],"responsible":[70],"for":[71],"most":[72],"PPO's":[74],"gain":[75],"cumulative":[77],"reward":[78],"over":[79],"TRPO,":[80],"(b)":[82],"fundamentally":[83],"change":[84],"how":[85],"RL":[86],"methods":[87],"function.":[88],"These":[89],"insights":[90],"difficulty,":[93],"attributing":[97],"performance":[98],"gains":[99],"reinforcement":[102],"learning.":[103]},"counts_by_year":[{"year":2025,"cited_by_count":19},{"year":2024,"cited_by_count":32},{"year":2023,"cited_by_count":15},{"year":2022,"cited_by_count":28},{"year":2021,"cited_by_count":39},{"year":2020,"cited_by_count":21}],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2019-12-26T00:00:00"}