%PDF-1.4%
182 0 obj<>/OCGs[185 0 R]>>/OutputIntents[179 0 R]/PageLabels 177 0 R/Pages 17 0 R/Type/Catalog>>endobj184 0 obj<>/Font<>>>/Fields 189 0 R>>endobj181 0 obj<>stream
2017-01-21T19:44:29-08:00
TeX
2017-01-22T21:56:16-08:00
2017-01-22T21:56:16-08:00
This is pdfTeX, Version 3.14159265-2.6-1.40.17 (TeX Live 2016) kpathsea version 6.2.2
Acrobat Distiller 15.0 (Macintosh)
False
Machine Learning Methods
application/pdf
Yi Xu, Haiqin Yang, Lijun Zhang, Tianbao Yang
Efficient Non-Oblivious Randomized Reduction for Risk Minimization with Improved Excess Risk Guarantee
Copyright© 2017 Association for the Advancement of Artificial Intelligence
Proceedings of the the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17)
Machine Learning Methods
True
http://www.aaai.org
uuid:63fdd739-9189-8a4c-95aa-93955c9590c5
uuid:0043c77a-61ad-f14d-8be6-b33c539438c2
1
B
endstreamendobj177 0 obj<>endobj17 0 obj<>endobj2 0 obj<>/Font<>/ProcSet[/PDF/Text]/XObject<>>>/Type/Page>>endobj18 0 obj<>/Font<>/ProcSet[/PDF/Text]/XObject<>>>/Type/Page>>endobj32 0 obj<>/Font<>/ProcSet[/PDF/Text]/XObject<>>>/Type/Page>>endobj37 0 obj<>/Font<>/ProcSet[/PDF/Text]/XObject<>>>/Type/Page>>endobj45 0 obj<>/Font<>/ProcSet[/PDF/Text]/XObject<>>>/Type/Page>>endobj49 0 obj<>/Font<>/ProcSet[/PDF/Text]/XObject<>>>/Type/Page>>endobj54 0 obj<>/Font<>/ProcSet[/PDF/Text]/XObject<>>>/Type/Page>>endobj208 0 obj<>stream
Hr_G0e`';NJlU*OrDۛ->Az*˵O-ZkBJ)"&,n}ۡ}W.]7},L$I UGFyeXeYVԶHši{ 2#VBZ9ڕQ I<_[4=Z6B>pMj_^%w=i#ݸaM%#۳=۞.|1Uմ,/Jw5~5>k#E~ce*Hx@Tf(bH9a7|ùo\[6vUϤ!uL= "Ϻ&mdChL ( t]:FMrG>]S?uQ:Y\)j-hjã$@ wR:mtE]PB}3UGR&@+#n$J
|5XQҠֆUDTB#vk{h1%"|Lj;Xמj)f)myR.'e
yw{|CR fEoj덻]@l~(T U.$24nϗ cLWT=rD&XE\/Äf|xHDr;182Q6ǔ7IHZD : ֣c.T,҈wT8$Ëwk:PCRCo-He BjkĤPʁÁ#O*_O<>P-P]aӕxGyel^NM"4s3V2a Ccd>
Ỿb$Pzq@Qe0BWL-h,[.LؑFq%VJw]Di&vn]7]nN&C ۑ.!͇G!!9 >jx㿨hI1fOD@=l-'aֶ||72-q*_{J~d>ʙ[\tv<40Mr!dɦ\P $q,&'&LpI|Dcv oQ]IL_h`[c*`"2&hfy8ժ06vL ಽ+CMSM
Ag!10A ,/v31bX摬 wO|zsb|5*?jI<t1)
Cs2#;1Bf*!c"=~܊T1Dݰz=r˟UQ[;(N1rwZυNRXRd6GvǴT75_"\< v=OhO~s$n;i]X r=C{S b7X`;aHZF#(6pK:jL6j"̭9$e #A[HF`IF*h3lp˖0ST-W/o6}x](Jw"1뽳 .;WC%ydңaq0Bd8Ղf-I,Ⱦij2l>\w|;ZOQ\@mQ뾅2z^A\ܱފY=f:,#Ձƚݼ;M-.:EyTc!gKo`&iVtc'R\G7kQ߆yev
/,Q:ۆ3Oyo_1 .lkw4?e> ;ڳu"GA3h"| ?)-8s#㒾[OِUD@h"yaݫ*xXs Pgl7B}h~Jqe PYu:?s4臡VR/uS^i1{,y{JxvĒ'ih )<34ƈ1$s*qvS^0c|BB_8U-*|evm@77gegﺗ@Jay꿮gqq>{f%ny&y@Bjo4,٣f.3 8h(HBn4iBHG$O^.*81vCUAF`##Huv
e#O2&~7`!6
HX廓Gy-mdw}*&)Ylٖ∪LHI@is:^l,ZbL"`_?K`pL`f$ XYIITNgcFv&EΖN8"~x1ਨt@WEE5LgV-t`4gh=wh}HF/tb1vmُ
dOm߫PÑOBC܃j&ƏJ@[o6oUS Vw}q1 J <pTcEMy}j,z켬&1ĺi
iPIȼ8z2=rcD~\.BF,d^IK߮krxn}ѷ"0?䕝"Gi<VEPvY571HZ%Z5/CُX H졀;OV6Qͭ3ՀeJ<<2vX*+>P8EI!~
B^Y!5d2o&0%(Iʟ,kň [Gb1Qr\N(?ֶEJ H2_V{}X\4kTAX!jeї,/CO)T=7]cKb3IU3a.C9gSzSVQ(_mtnS(:/OrZR#@d
Lb&yhّ`P\*Mtǒt˖ m]{.SOrrHO]vK^2$[rZ^ZЩ:d~PK^0"q)4AqG+pb*#t+yC6ۺ('dtV@G*tM2뿨b.cO]{^yG/bG(H"nŬפǣ,fMoLz]m*Rf&dUa'84-S'aĐhx릨e49ֆ;=S8Q=؝.uňKBqGҁwq;|3snԜ)mhGj?Q%}&ӀĢ_+}عsM(kU]*!\GG&1~"1'N {;ևbUPɭꌔ>Z93F68<ęDXN@d
yG#|hIimH,i\Pʒt$o"]c^IqW XVvQgC\~s*eya
EBVȷ-Hs7KQu|L!:U