%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\054 Inc\056) /Language (en\055US) /Created (2016) /EventType (Poster) /Description-Abstract (We study the problem of completing a binary matrix in an online learning setting\056 On each trial we predict a matrix entry and then receive the true entry\056 We propose a Matrix Exponentiated Gradient algorithm \1331\135 to solve this problem\056 We provide a mistake bound for the algorithm\054 which scales with the margin complexity \1332\054 3\135 of the underlying matrix\056 The bound suggests an interpretation where each row of the matrix is a prediction task over a finite set of objects\054 the columns\056 Using this we show that the algorithm makes a number of mistakes which is comparable up to a logarithmic factor to the number of mistakes made by the Kernel Perceptron with an optimal kernel in hindsight\056 We discuss applications of the algorithm to predicting as well as the best biclustering and to the problem of predicting the labeling of a graph without knowing the graph in advance\056) /Producer (PyPDF2) /Title (Mistake Bounds for Binary Matrix Completion) /Date (2016) /ModDate (D\07220170112154328\05508\04700\047) /Published (2016) /Type (Conference Proceedings) /firstpage (3954) /Book (Advances in Neural Information Processing Systems 29) /Description (Paper accepted and presented at the Neural Information Processing Systems Conference \050http\072\057\057nips\056cc\057\051) /Editors (D\056D\056 Lee and M\056 Sugiyama and U\056V\056 Luxburg and I\056 Guyon and R\056 Garnett) /Author (Mark Herbster\054 Stephen Pasteris\054 Massimiliano Pontil) /lastpage (3962) >> 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 0 612 792 ] >> endobj 5 0 obj << /Parent 1 0 R /Contents 95 0 R /Type /Page /Resources 96 0 R /MediaBox [ 0 0 612 792 ] >> endobj 6 0 obj << /Parent 1 0 R /Contents 157 0 R /Type /Page /Resources 158 0 R /MediaBox [ 0 0 612 792 ] >> endobj 7 0 obj << /Parent 1 0 R /Contents 177 0 R /Type /Page /Resources 178 0 R /MediaBox [ 0 0 612 792 ] >> endobj 8 0 obj << /Parent 1 0 R /Contents 193 0 R /Type /Page /Resources 194 0 R /MediaBox [ 0 0 612 792 ] >> endobj 9 0 obj << /Parent 1 0 R /Contents 202 0 R /Type /Page /Resources 203 0 R /MediaBox [ 0 0 612 792 ] >> endobj 10 0 obj << /Parent 1 0 R /Contents 204 0 R /Type /Page /Resources 205 0 R /MediaBox [ 0 0 612 792 ] >> endobj 11 0 obj << /Parent 1 0 R /Contents 212 0 R /Type /Page /Resources 213 0 R /MediaBox [ 0 0 612 792 ] >> endobj 12 0 obj << /Parent 1 0 R /Contents 214 0 R /Type /Page /Resources 215 0 R /MediaBox [ 0 0 612 792 ] >> endobj 13 0 obj << /Length 5220 /Filter /FlateDecode >> stream x\Kqlz" Px4-mJTlӃ 4ZU?Cg_G3{X4P/Gޏ^ya%yei*[0KԏS{QJRvz/H 6;[/|&40R~(/ry;ۏz?xo]xT_y7Ju{/</~[궹[-RA1fqYzXW"Y{#̺
@>G#9/A?ClfpUL"IALPZT+^7ZKf
=*XH'TY4;t>Sa"UqY4Hzt|Dq' f NU&г쓑NoR:((*Go
]/OF~^fd#<]˦R6z|UP^@Z9}|}<էTj*þj4}3jsU[UiVx d5IiFv_WM[:E`gf<@?벹nuRЯ}(,i74*C/#5b! V^'~,h|!Pز8KA,hZ(5ZWOJ0"$в* Gd1%t(7K(T2e1MS/?f{ƅ>(؈C{v=pic`g`۠ΡUv9_\>mufV塩a鄟rϊHE}j`;
ͣ l8%->h7A
\sPb>