%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 ] /Type /Pages /Count 9 >> endobj 2 0 obj << /Subject (Neural Information Processing Systems http\072\057\057nips\056cc\057) /Publisher (Curran Associates) /Language (en\055US) /Created (2010) /Description-Abstract (In multi\055instance learning\054 there are two kinds of prediction failure\054 i\056e\056\054 false negative and false positive\056 Current research mainly focus on avoding the former\056 We attempt to utilize the geometric distribution of instances inside positive bags to avoid both the former and the latter\056 Based on kernel principal component analysis\054 we define a projection constraint for each positive bag to classify its constituent instances far away from the separating hyperplane while place positive instances and negative instances at opposite sides\056 We apply the Constrained Concave\055Convex Procedure to solve the resulted problem\056 Empirical results demonstrate that our approach offers improved generalization performance\056) /Producer (Python PDF Library \055 http\072\057\057pybrary\056net\057pyPdf\057) /Title (Avoiding False Positive in Multi\055Instance Learning) /Date (2010) /Type (Conference Proceedings) /firstpage (811) /Book (Advances in Neural Information Processing Systems 23) /Description (Paper accepted and presented at the Neural Information Processing Systems Conference \050http\072\057\057nips\056cc\057\051) /Editors (J\056D\056 Lafferty and C\056K\056I\056 Williams and J\056 Shawe\055Taylor and R\056S\056 Zemel and A\056 Culotta) /Author (Yanjun Han\054 Qing Tao\054 Jue Wang) /lastpage (819) >> endobj 3 0 obj << /Type /Catalog /Pages 1 0 R >> endobj 4 0 obj << /Parent 1 0 R /Contents [ 13 0 R ] /Type /Page /Resources 14 0 R /MediaBox [ 0 0 595.28000 841.89000 ] >> endobj 5 0 obj << /Parent 1 0 R /Contents [ 38 0 R ] /Type /Page /Resources 39 0 R /MediaBox [ 0 0 595.28000 841.89000 ] >> endobj 6 0 obj << /Parent 1 0 R /Contents [ 69 0 R ] /Type /Page /Resources 70 0 R /MediaBox [ 0 0 595.28000 841.89000 ] >> endobj 7 0 obj << /Parent 1 0 R /Contents [ 97 0 R ] /Type /Page /Resources 98 0 R /MediaBox [ 0 0 595.28000 841.89000 ] >> endobj 8 0 obj << /Parent 1 0 R /Contents [ 99 0 R ] /Type /Page /Resources 100 0 R /MediaBox [ 0 0 595.28000 841.89000 ] >> endobj 9 0 obj << /Parent 1 0 R /Contents [ 124 0 R ] /Type /Page /Resources 125 0 R /MediaBox [ 0 0 595.28000 841.89000 ] >> endobj 10 0 obj << /Parent 1 0 R /Contents [ 126 0 R ] /Type /Page /Resources 127 0 R /MediaBox [ 0 0 595.28000 841.89000 ] >> endobj 11 0 obj << /Parent 1 0 R /Contents [ 128 0 R ] /Type /Page /Resources 129 0 R /MediaBox [ 0 0 595.28000 841.89000 ] >> endobj 12 0 obj << /Parent 1 0 R /Contents [ 130 0 R ] /Type /Page /Resources 131 0 R /MediaBox [ 0 0 595.28000 841.89000 ] >> endobj 13 0 obj << /Length 3033 /Filter /FlateDecode >> stream xڍYK8QJku۞Ld*ڭN̶Ȓ#Jq:~> %?f+ Exu/"e(*\Uz^}.ҰZ"-wqFbOŇ?q&qxxZdeX< xyq7ƴ]UeZqtۭ< ލ`vPmV2~zucp9c/LªHIHü",V,
8Lȷ5`rqD7Z{6s260VΝe0tV%.,vcLjSM 4cgPnkdthV/<.aU$';,lhM>SY9$˱u;NP}2mxuheȢӾ-t)zBK/Ƴ2e E+>zhC'q0'Vw{=͚vEďT cj6ϥZ]-2X$IEei
;Yd i I\6tV!#=EvH$>}uF8A5Zw=t\'HyҼ
t30ۍEIfJ>ab^w-y^%_'@+"To
u2i95Ԛ'9%/dO/>@g;HQ