{"id":"https://openalex.org/W6987609456","doi":null,"title":"Towards Responsible Machine Learning: Enhancing Robustness Against Adversaries and Corruptions and Promoting Fairness in Decision-Making","display_name":"Towards Responsible Machine Learning: Enhancing Robustness Against Adversaries and Corruptions and Promoting Fairness in Decision-Making","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W6987609456"},"language":"en","primary_location":{"id":"pmh:oai:escholarship.org:ark:/13030/qt7q93b81v","doi":null,"is_oa":true,"landing_page_url":"https://escholarship.org/uc/item/7q93b81v","pdf_url":"https://escholarship.org/content/qt7q93b81v/qt7q93b81v.pdf","source":{"id":"https://openalex.org/S4306400115","display_name":"eScholarship (California Digital Library)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2801248553","host_organization_name":"California Digital Library","host_organization_lineage":["https://openalex.org/I2801248553"],"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":"etd"},"volume":null,"type":"dissertation","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://escholarship.org/content/qt7q93b81v/qt7q93b81v.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5063721204","display_name":"Bhagyashree Puranik","orcid":"https://orcid.org/0000-0003-1083-137X","observed_orcids":["https://orcid.org/0000-0003-1083-137X"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Puranik, Bhagyashree","raw_affiliation_strings":["Education 2024"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Education 2024","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5063721204"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"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":true,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9961000084877014,"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":0.9961000084877014,"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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.00139999995008111,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.0006000000284984708,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/robust-machine-learning","display_name":"robust machine learning","score":0.7929999828338623},{"id":"https://openalex.org/keywords/adversarial-perturbations","display_name":"adversarial perturbations","score":0.7990000247955322},{"id":"https://openalex.org/keywords/generalized-likelihood-ratio-test","display_name":"generalized likelihood ratio test","score":0.7580000162124634},{"id":"https://openalex.org/keywords/deep-neural-network-robustness","display_name":"deep neural network robustness","score":0.8399999737739563},{"id":"https://openalex.org/keywords/positive-reinforcement","display_name":"positive reinforcement","score":0.4059999883174896},{"id":"https://openalex.org/keywords/responsible-machine-learning","display_name":"responsible machine learning","score":0.5059999823570251}],"concepts":[{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.8205999732017517},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.7646999955177307},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7170000076293945},{"id":"https://openalex.org/C149728462","wikidata":"https://www.wikidata.org/wiki/Q751319","display_name":"Minimax","level":2,"score":0.6341000199317932},{"id":"https://openalex.org/C153701036","wikidata":"https://www.wikidata.org/wiki/Q659974","display_name":"Trustworthiness","level":2,"score":0.5636000037193298},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.4672999978065491},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.44909998774528503},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42739999294281006},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.4081000089645386},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.36739999055862427},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.367000013589859},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.36010000109672546},{"id":"https://openalex.org/C140547941","wikidata":"https://www.wikidata.org/wiki/Q7797194","display_name":"Threat model","level":2,"score":0.3458999991416931},{"id":"https://openalex.org/C87007009","wikidata":"https://www.wikidata.org/wiki/Q210832","display_name":"Statistical hypothesis testing","level":2,"score":0.2865000069141388},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.2628999948501587},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.2621000111103058},{"id":"https://openalex.org/C79772020","wikidata":"https://www.wikidata.org/wiki/Q5159264","display_name":"Conditional independence","level":2,"score":0.2549999952316284},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.2526000142097473}],"mesh":[],"locations_count":1,"locations":[{"id":"pmh:oai:escholarship.org:ark:/13030/qt7q93b81v","doi":null,"is_oa":true,"landing_page_url":"https://escholarship.org/uc/item/7q93b81v","pdf_url":"https://escholarship.org/content/qt7q93b81v/qt7q93b81v.pdf","source":{"id":"https://openalex.org/S4306400115","display_name":"eScholarship (California Digital Library)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2801248553","host_organization_name":"California Digital Library","host_organization_lineage":["https://openalex.org/I2801248553"],"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":"etd"}],"best_oa_location":{"id":"pmh:oai:escholarship.org:ark:/13030/qt7q93b81v","doi":null,"is_oa":true,"landing_page_url":"https://escholarship.org/uc/item/7q93b81v","pdf_url":"https://escholarship.org/content/qt7q93b81v/qt7q93b81v.pdf","source":{"id":"https://openalex.org/S4306400115","display_name":"eScholarship (California Digital Library)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2801248553","host_organization_name":"California Digital Library","host_organization_lineage":["https://openalex.org/I2801248553"],"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":"etd"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320322724","display_name":"Ministry of Education, India","ror":"https://ror.org/048xjjh50"},{"id":"https://openalex.org/F4320332500","display_name":"University of California, Santa Barbara","ror":"https://ror.org/02t274463"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W6987609456.pdf","grobid_xml":"https://content.openalex.org/works/W6987609456.grobid-xml"},"referenced_works_count":14,"referenced_works":["https://openalex.org/W1686810756","https://openalex.org/W2787733970","https://openalex.org/W2810134417","https://openalex.org/W2896457183","https://openalex.org/W2946948417","https://openalex.org/W2964121744","https://openalex.org/W2964137095","https://openalex.org/W2970247852","https://openalex.org/W2985461281","https://openalex.org/W3030163527","https://openalex.org/W3035656068","https://openalex.org/W3093235747","https://openalex.org/W4280498404","https://openalex.org/W4287328355"],"related_works":["https://openalex.org/W6943734735","https://openalex.org/W7162476999","https://openalex.org/W7081678711","https://openalex.org/W6986513130","https://openalex.org/W7043510665","https://openalex.org/W7053013795","https://openalex.org/W6907913894","https://openalex.org/W7033745423","https://openalex.org/W7132995918","https://openalex.org/W4283397204"],"abstract_inverted_index":{"Designing":[0],"trustworthy":[1],"machine":[2],"learning":[3,190],"(ML)":[4],"systems":[5,379],"requires":[6],"addressing":[7],"critical":[8],"challenges":[9,348],"in":[10,61,67,131,153,212,241,258,266,312,349],"robustness":[11,60,166,199,343],"and":[12,22,36,64,95,113,150,162,170,197,219,226,281,295,340,344,358,367],"fairness.":[13,323],"Robust":[14],"ML":[15,62,378],"models":[16,63],"must":[17,224],"withstand":[18],"corruptions,":[19],"adversarial":[20,75,88,151],"perturbations":[21,89,152],"other":[23],"uncertainties":[24],"to":[25,33,58,106,122,146,285,332,335,337,370],"ensure":[26],"reliable":[27,357],"performance,":[28],"while":[29,125],"fairness":[30,66,211,223,232,277,286,345],"mechanisms":[31],"aim":[32],"mitigate":[34],"biases":[35,240],"promote":[37],"equitable":[38],"outcomes":[39],"across":[40],"diverse":[41],"scenarios.":[42],"This":[43,116,362],"thesis":[44],"explores":[45],"these":[46,372],"two":[47],"distinct":[48,347],"but":[49,302],"vital":[50],"dimensions":[51],"of":[52,111,308,377],"responsible":[53,350,360],"ML,":[54,351],"presenting":[55],"novel":[56,144],"approaches":[57],"enhance":[59,165],"long-term":[65],"decision-making":[68,214,235],"systems.We":[69],"first":[70],"investigate":[71],"fundamental":[72],"defenses":[73,124],"against":[74],"attacks":[76],"by":[77,167],"studying":[78],"a":[79,98,127,143,206],"simplified":[80],"Gaussian":[81],"hypothesis":[82,252],"testing":[83],"scenario.":[84],"In":[85,205],"this":[86],"setting,":[87],"are":[90,353],"modeled":[91],"as":[92,217],"nuisance":[93],"parameters,":[94],"we":[96,141,209,303],"propose":[97,270],"defense":[99],"using":[100],"the":[101,109,114,175,251,271,306,327,375],"generalized":[102],"likelihood":[103],"ratio":[104,177],"test":[105],"jointly":[107],"estimate":[108],"class":[110],"interest":[112],"perturbation.":[115],"approach":[117,145],"achieves":[118,198],"asymptotic":[119],"performance":[120],"comparable":[121],"minimax":[123],"offering":[126],"favorable":[128],"robustness-accuracy":[129],"trade-off":[130],"non-asymptotic":[132],"regimes.":[133],"Next,":[134],"building":[135],"on":[136,184,202,293],"insights":[137,366],"from":[138],"communication":[139],"theory,":[140],"introduce":[142],"combating":[147],"out-of-distribution":[148],"corruptions":[149],"deep":[154],"neural":[155],"networks.":[156],"Our":[157],"tilted":[158],"exponential":[159],"(TEXP)":[160],"layer":[161],"training":[163],"objective":[164],"promoting":[168],"sparse":[169],"meaningful":[171],"layer-wise":[172],"representations,":[173],"improving":[174],"signal-to-noise":[176],"during":[178],"learning.":[179],"We":[180,249,269],"show":[181],"through":[182],"experiments":[183],"standard":[185,193],"image":[186],"datasets":[187,298],"that":[188,253],"TEXP":[189],"integrates":[191],"with":[192,278],"empirical":[194],"risk":[195],"minimization":[196],"without":[200],"relying":[201],"data":[203],"augmentation.":[204],"complementary":[207],"study,":[208],"address":[210,346],"sequential":[213,234],"processes,":[215],"such":[216],"hiring":[218],"college":[220],"admissions,":[221],"where":[222,239,317],"evolve":[225],"persist":[227],"over":[228],"time.":[229],"Unlike":[230],"one-shot":[231],"metrics,":[233],"introduces":[236],"feedback":[237,319],"loops":[238,320],"early":[242],"decisions":[243],"can":[244,260],"influence":[245],"future":[246],"applicant":[247],"pools.":[248],"examine":[250],"slightly":[254],"favoring":[255],"underrepresented":[256],"groups":[257],"selection":[259],"generate":[261],"positive":[262],"reinforcement,":[263],"increasing":[264],"diversity":[265],"subsequent":[267],"rounds.":[268],"Multi-agent":[272],"Fair-Greedy":[273],"policy,":[274],"which":[275],"balances":[276],"score":[279],"maximization,":[280],"prove":[282],"its":[283,300],"convergence":[284],"targets":[287],"under":[288],"certain":[289],"conditions.":[290],"Empirical":[291],"results":[292],"synthetic":[294],"adapted":[296],"real-world":[297,381],"demonstrate":[299],"effectiveness,":[301],"also":[304],"highlight":[305],"risks":[307],"uncoordinated":[309],"agent":[310],"behavior":[311],"more":[313],"complex":[314],"population":[315],"dynamics,":[316],"negative":[318],"may":[321],"reduce":[322],"These":[324],"findings":[325],"emphasize":[326],"need":[328],"for":[329,355,380],"policy":[330],"makers":[331],"design":[333],"carefully":[334],"respond":[336],"evolving":[338],"social":[339],"demographic":[341],"conditions.Although":[342],"both":[352],"essential":[354],"creating":[356],"socially":[359],"systems.":[361],"work":[363],"provides":[364],"foundational":[365],"practical":[368],"tools":[369],"tackle":[371],"dimensions,":[373],"advancing":[374],"development":[376],"applications.":[382]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2025-10-10T00:00:00"}