%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 (2011) /Description-Abstract (We consider an adversarial online learning setting where a decision maker can choose an action in every stage of the game\056 In addition to observing the reward of the chosen action\054 the decision maker gets side observations on the reward he would have obtained had he chosen some of the other actions\056 The observation structure is encoded as a graph\054 where node i is linked to node j if sampling i provides information on the reward of j\056 This setting naturally interpolates between the well\055known \140\140experts\047\047 setting\054 where the decision maker can view all rewards\054 and the multi\055armed bandits setting\054 where the decision maker can only view the reward of the chosen action\056 We develop practical algorithms with provable regret guarantees\054 which depend on non\055trivial graph\055theoretic properties of the information feedback structure\056 We also provide partially\055matching lower bounds\056) /Producer (Python PDF Library \055 http\072\057\057pybrary\056net\057pyPdf\057) /Title (From Bandits to Experts\072 On the Value of Side\055Observations) /Date (2011) /Type (Conference Proceedings) /firstpage (684) /Book (Advances in Neural Information Processing Systems 24) /Description (Paper accepted and presented at the Neural Information Processing Systems Conference \050http\072\057\057nips\056cc\057\051) /Editors (J\056 Shawe\055Taylor and R\056S\056 Zemel and P\056L\056 Bartlett and F\056 Pereira and K\056Q\056 Weinberger) /Author (Shie Mannor\054 Ohad Shamir) /lastpage (692) >> 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 56 0 R /Type /Page /Resources 57 0 R /MediaBox [ 0 0 612 792 ] >> endobj 6 0 obj << /Parent 1 0 R /Contents 70 0 R /Type /Page /Resources 71 0 R /MediaBox [ 0 0 612 792 ] >> endobj 7 0 obj << /Parent 1 0 R /Contents 80 0 R /Type /Page /Resources 81 0 R /MediaBox [ 0 0 612 792 ] >> endobj 8 0 obj << /Parent 1 0 R /Contents 86 0 R /Type /Page /Resources 87 0 R /MediaBox [ 0 0 612 792 ] >> endobj 9 0 obj << /Parent 1 0 R /Contents 100 0 R /Type /Page /Resources 101 0 R /MediaBox [ 0 0 612 792 ] >> endobj 10 0 obj << /Parent 1 0 R /Contents 106 0 R /Type /Page /Resources 107 0 R /MediaBox [ 0 0 612 792 ] >> endobj 11 0 obj << /Parent 1 0 R /Contents 112 0 R /Type /Page /Resources 113 0 R /MediaBox [ 0 0 612 792 ] >> endobj 12 0 obj << /Parent 1 0 R /Contents 126 0 R /Type /Page /Resources 127 0 R /MediaBox [ 0 0 612 792 ] >> endobj 13 0 obj << /Length 3380 /Filter /FlateDecode >> stream xڍZKБv @HvlZTkD IdUW5Ii/wuu=*N{E~N7ک]Vaƻ}Gq,`wwxӻ?|;ZSR힎oU}i|mc=ܘz;L-~(u`4O,Lu۫2,ug}v0Z菙iŻoatvG|W}uZwh>n3*r`ĹNfl+WI 8--yݟSki'K4ԕi=םC=;A$:,TyC:w#/nm4Z3)qrK|em8&ZV߈@Wn<ȇ~<ޔ@\i? iGk̭hiz'}|uӤ۩=_찂{*vqN@L uqL59+Ǘ8^'a?<x)h~4
BNKbOjHA}8n-cYm4.FG2cVbu=Sq"{u}`
&: AG"Y㞐aGǾN;i=(z}ힴƛB*TgI8xqkõtZA=HRH