%PDF-1.3
1 0 obj
<<
/Kids [ 4 0 R 5 0 R 6 0 R 7 0 R 8 0 R 9 0 R 10 0 R 11 0 R 12 0 R 13 0 R ]
/Type /Pages
/Count 10
>>
endobj
2 0 obj
<<
/Subject (Neural Information Processing Systems http\072\057\057nips\056cc\057)
/Publisher (Curran Associates\054 Inc\056)
/Language (en\055US)
/Created (2017)
/EventType (Poster)
/Description-Abstract (Machine learning is now being used to make crucial decisions about people\047s lives\056 For nearly all of these decisions there is a risk that individuals of a certain race\054 gender\054 sexual orientation\054 or any other subpopulation are unfairly discriminated against\056 Our recent method has demonstrated how to use techniques from counterfactual inference to make predictions fair across different subpopulations\056 This method requires that one provides the causal model that generated the data at hand\056 In general\054 validating all causal implications of the model is not possible without further assumptions\056 Hence\054 it is desirable to integrate competing causal models to provide counterfactually fair decisions\054 regardless of which causal \042world\042 is the correct one\056 In this paper\054 we show how it is possible to make predictions that are approximately fair with respect to multiple possible causal models at once\054 thus mitigating the problem of exact causal specification\056 We frame the goal of learning a fair classifier as an optimization problem with fairness constraints entailed by competing causal explanations\056 We show how this optimization problem can be efficiently solved using gradient\055based methods\056 We demonstrate the flexibility of our model on two real\055world fair classification problems\056 We show that our model can seamlessly balance fairness in multiple worlds with prediction accuracy\056)
/Producer (PyPDF2)
/Title (When Worlds Collide\072 Integrating Different Counterfactual Assumptions in Fairness)
/Date (2017)
/ModDate (D\07220180212210846\05508\04700\047)
/Published (2017)
/Type (Conference Proceedings)
/firstpage (6414)
/Book (Advances in Neural Information Processing Systems 30)
/Description (Paper accepted and presented at the Neural Information Processing Systems Conference \050http\072\057\057nips\056cc\057\051)
/Editors (I\056 Guyon and U\056V\056 Luxburg and S\056 Bengio and H\056 Wallach and R\056 Fergus and S\056 Vishwanathan and R\056 Garnett)
/Author (Chris Russell\054 Matt J\056 Kusner\054 Joshua Loftus\054 Ricardo Silva)
/lastpage (6423)
>>
endobj
3 0 obj
<<
/Type /Catalog
/Pages 1 0 R
>>
endobj
4 0 obj
<<
/Contents 14 0 R
/Parent 1 0 R
/Resources 15 0 R
/MediaBox [ 0 0 612 792 ]
/Annots [ 38 0 R 39 0 R 40 0 R 41 0 R 42 0 R 43 0 R 44 0 R 45 0 R ]
/Type /Page
>>
endobj
5 0 obj
<<
/Contents 46 0 R
/Parent 1 0 R
/Resources 47 0 R
/MediaBox [ 0 0 612 792 ]
/Annots [ 64 0 R 65 0 R 66 0 R 67 0 R 68 0 R 69 0 R 70 0 R 71 0 R 72 0 R 73 0 R 74 0 R 75 0 R 76 0 R 77 0 R 78 0 R 79 0 R ]
/Type /Page
>>
endobj
6 0 obj
<<
/Contents 80 0 R
/Parent 1 0 R
/Resources 81 0 R
/MediaBox [ 0 0 612 792 ]
/Annots [ 121 0 R 122 0 R ]
/Type /Page
>>
endobj
7 0 obj
<<
/Contents 123 0 R
/Parent 1 0 R
/Resources 124 0 R
/MediaBox [ 0 0 612 792 ]
/Annots [ 145 0 R 146 0 R 147 0 R 148 0 R 149 0 R 150 0 R 151 0 R 152 0 R 153 0 R 154 0 R 155 0 R ]
/Type /Page
>>
endobj
8 0 obj
<<
/Contents 156 0 R
/Parent 1 0 R
/Resources 157 0 R
/MediaBox [ 0 0 612 792 ]
/Annots [ 162 0 R 163 0 R 164 0 R 165 0 R 166 0 R 167 0 R 168 0 R ]
/Type /Page
>>
endobj
9 0 obj
<<
/Contents 169 0 R
/Parent 1 0 R
/Resources 170 0 R
/MediaBox [ 0 0 612 792 ]
/Annots [ 179 0 R 180 0 R 181 0 R ]
/Type /Page
>>
endobj
10 0 obj
<<
/Contents 182 0 R
/Parent 1 0 R
/Resources 183 0 R
/MediaBox [ 0 0 612 792 ]
/Annots [ 184 0 R 185 0 R 186 0 R 187 0 R 188 0 R 189 0 R ]
/Type /Page
>>
endobj
11 0 obj
<<
/Contents 190 0 R
/Parent 1 0 R
/Resources 191 0 R
/MediaBox [ 0 0 612 792 ]
/Annots [ 308 0 R 309 0 R 310 0 R 311 0 R 312 0 R 313 0 R 314 0 R 315 0 R 316 0 R ]
/Type /Page
>>
endobj
12 0 obj
<<
/Contents 317 0 R
/Parent 1 0 R
/Resources 318 0 R
/MediaBox [ 0 0 612 792 ]
/Annots [ 371 0 R 372 0 R 373 0 R 374 0 R 375 0 R ]
/Type /Page
>>
endobj
13 0 obj
<<
/Contents 376 0 R
/Parent 1 0 R
/Resources 377 0 R
/MediaBox [ 0 0 612 792 ]
/Annots [ 378 0 R 379 0 R 380 0 R 381 0 R 382 0 R 383 0 R 384 0 R 385 0 R 386 0 R 387 0 R 388 0 R 389 0 R 390 0 R 391 0 R 392 0 R 393 0 R 394 0 R 395 0 R 396 0 R 397 0 R 398 0 R 399 0 R 400 0 R 401 0 R 402 0 R 403 0 R 404 0 R 405 0 R 406 0 R ]
/Type /Page
>>
endobj
14 0 obj
<<
/Length 3058
/Filter /FlateDecode
>>
stream
xÚYK“¤Æ¾ï¯è›èˆà“V²ÖÚ•,ËÚQ(’4TOW^šÿzU™EÝë>A½3³2¿|T°{Ü»¿½
øûáUˆo°waíÒ ó³0Ù•Í+Ó)r¹~ž%»^펯þùêˇWŸ¿‘˜›úQ˜Ä»‡ã.Œ…Ÿ&ÉN¦™Ÿ'b÷Pí~ñ~>©v%÷ó>M¼®¯«Ú_uu+õ—ý½ïm;ªÇ¾uûHÃÕÇ£ê÷aæ©vÜÿöðûûIîîÃÜÏ£„¶ÿª›°°?å85|=Ssu×òIš)x³¯ÐØ4ñZ5f×W_?l8—"÷¥°œÿò[°«0ðnø9$ðl§5VÁ®Þ½_DÉHZ9$‰ŸÆb'áG2eBO½fr~œ†AÕµ9ýó7!æùy*2³2‘~*Œ e(iÙ¯A h¢‹#î3ÜŽ•dâdñpR,€º`ŽöqâMý,Ô·í0êqybÑV$ÙÌOr³ØrÚí§Vqýn„¥úA/´¦;Ò÷ýÔ÷ʾ}"XÒº¢(_ïXöÄø£¥È/JÚ‡Þó·¡Œ|¨ÑND~±ÿ^Œ#ýnžOÿߦ¡…ª\ËSÆØ%ý#òÌC_òÿ&O˜
(øSòüyŸaƒþY—O·*ÍïzÇæÉò½çMYBŸte)Ëw†Šn8M«$Kó»î8NÃ
I¦‘dùR’/·ÄËb#Æï•áû™ø×>„±wý5oåš÷Ä—ÙF˜µ¥ó‹öeòU5Ý`ú>»tî,·Ë~ÔeÑW+°®q.d~‹øÆÈW‡þïúøA$ÿ˜:T´fõ¨Ü•´U×ÞR‰ÄÏòx½kO~1ŒÅ8øSY€-#ÎIdv)v»û8ôÃ,¢^†±šZl6.b#%œ ‡œ ÿLÊ]Ç~CÞ5Ùò¤[ËHâÕªè[–”ôÃݬ‰wPóø4¨Š:ÇŽzšâÉ\ï0)5!¾ô*UêÁá= þÐM#œUw®Õg{üò`íä.=5ød®¸ž-Ø-rbãµ–ëÓŽ£èëú/jC@*ȈÑ1žÔ èwIôŠ&kî*è¿ðäæ#Oi+{¿ë
nçÓ!Â+U?Öµ™MŠRÝ9fà¶7ÜD0AÌ<ª¶XŠÜ»ÃR¨ï`Õí#¹ÏØ0ªáqã?íI¬cZoñž:
G¼Ãt8w究kxªeÖG{´F¦IhXXé¡ìu£ÛbTÕ'å5£H.÷Æoö S"‰½Lædz½*Mt€ÿÈkÔxê*ê?uVªÁ@£G«Nè9Í
‡†Õ-ÌŸì™UžZýaR
û®¡‘‘Æ=‘»¥5ü$ŒˆVŒX†]@ç¸MD2ª-Ù¢í¡è^*tà{UéÒE-vrc()ûnà‘J
vSv‰«+|Š©NÚb¹5Ô„E†0'`¥˜eéÕ‡I÷ŠÍ…uöÓ9C>÷Ñè&4‡ÌJýfÃ,¦ì2ñš®Rõv( Â}`)YøbmUŒ›ðHß t6R ܆ø ´<¼µª—ñþõÉêwó!Ðt¡%fÙâg¦Õ\RŠÎ5 tqÖä0‘(Äã¿šç´ÝHg\>Ô<óYC¤Ó8Ëk“B¦~Š^KûqêÙœØ÷%~µ{ßå¹£Q=ò×
Þhø û‚ŽE7¡¥A8!`œ‰©y¨ìš³r¶X\” ¶xCìðIy#@ÍRI4/Õ §ÖBÖêO&Ÿ.Uï¨Ma¤µñ¾ªiyʲO½ç“.O|Â|_©÷k‰ç½
&ê
˜º5¯67vËXEœÀ¥ÄBÙ!†-@C£ßVÜ!izÆ zQœq;q`1æö¬¨¸ Z‹FÀ÷zóíp€Þÿ†–‚‚·1€ú ¬ñEÂ.·8ã^>ê·n¥Ÿd+ !½¤?Øø™¹Ï˜ (øTúLd®T:Yˆ?É“±±f i©4„OèÙÒ™}k£Gm¯Þ)fâà@
ÃÙ
õRœ§m´úcQ2š,áÆ°£áGdÃö6SX‡.j8öE£6 óعõîˆ9Rù”ÀÃEp`½q²2Ze
#¶¤(î±)µ>Ô~;x£ÿí\fš,¸Eƒoi±qK†a6'çã®umÑÃË'é.aÀ
–b´F