%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 14 0 R 15 0 R ] /Type /Pages /Count 12 >> 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 (2019) /EventType (Poster) /Description-Abstract (Bayesian optimization has recently emerged as a popular method for the sample\055efficient optimization of expensive black\055box functions\056 However\054 the application to high\055dimensional problems with several thousand observations remains challenging\054 and on difficult problems Bayesian optimization is often not competitive with other paradigms\056 In this paper we take the view that this is due to the implicit homogeneity of the global probabilistic models and an overemphasized exploration that results from global acquisition\056 This motivates the design of a local probabilistic approach for global optimization of large\055scale high\055dimensional problems\056 We propose the TuRBO algorithm that fits a collection of local models and performs a principled global allocation of samples across these models via an implicit bandit approach\056 A comprehensive evaluation demonstrates that TuRBO outperforms state\055of\055the\055art methods from machine learning and operations research on problems spanning reinforcement learning\054 robotics\054 and the natural sciences\056) /Producer (PyPDF2) /Title (Scalable Global Optimization via Local Bayesian Optimization) /Date (2019) /ModDate (D\07220200213000309\05508\04700\047) /Published (2019) /Type (Conference Proceedings) /firstpage (5496) /Book (Advances in Neural Information Processing Systems 32) /Description (Paper accepted and presented at the Neural Information Processing Systems Conference \050http\072\057\057nips\056cc\057\051) /Editors (H\056 Wallach and H\056 Larochelle and A\056 Beygelzimer and F\056 d\047Alch\351\055Buc and E\056 Fox and R\056 Garnett) /Author (David Eriksson\054 Michael Pearce\054 Jacob Gardner\054 Ryan D\056 Turner\054 Matthias Poloczek) /lastpage (5507) >> endobj 3 0 obj << /Type /Catalog /Pages 1 0 R >> endobj 4 0 obj << /Contents 16 0 R /Parent 1 0 R /Resources 17 0 R /MediaBox [ 0 0 612 792 ] /Annots [ 39 0 R 40 0 R 41 0 R 42 0 R 43 0 R 44 0 R 45 0 R 46 0 R 47 0 R 48 0 R ] /Type /Page >> endobj 5 0 obj << /Contents 49 0 R /Parent 1 0 R /Resources 50 0 R /MediaBox [ 0 0 612 792 ] /Annots [ 51 0 R 52 0 R 53 0 R 54 0 R 55 0 R 56 0 R 57 0 R 58 0 R 59 0 R 60 0 R 61 0 R 62 0 R 63 0 R 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 80 0 R 81 0 R 82 0 R 83 0 R 84 0 R 85 0 R 86 0 R 87 0 R 88 0 R 89 0 R 90 0 R 91 0 R 92 0 R 93 0 R 94 0 R ] /Type /Page >> endobj 6 0 obj << /Contents 95 0 R /Parent 1 0 R /Resources 96 0 R /MediaBox [ 0 0 612 792 ] /Annots [ 133 0 R 134 0 R 135 0 R 136 0 R 137 0 R 138 0 R 139 0 R 140 0 R 141 0 R 142 0 R 143 0 R 144 0 R 145 0 R 146 0 R ] /Type /Page >> endobj 7 0 obj << /Contents 147 0 R /Parent 1 0 R /Resources 148 0 R /MediaBox [ 0 0 612 792 ] /Annots [ 157 0 R 158 0 R 159 0 R ] /Type /Page >> endobj 8 0 obj << /Contents 160 0 R /Parent 1 0 R /Resources 161 0 R /MediaBox [ 0 0 612 792 ] /Annots [ 172 0 R 173 0 R 174 0 R 175 0 R 176 0 R 177 0 R ] /Type /Page >> endobj 9 0 obj << /Contents 178 0 R /Parent 1 0 R /Resources 179 0 R /Group 211 0 R /MediaBox [ 0 0 612 792 ] /Annots [ 217 0 R 218 0 R 219 0 R 220 0 R 221 0 R 222 0 R 223 0 R 224 0 R ] /Type /Page >> endobj 10 0 obj << /Contents 225 0 R /Parent 1 0 R /Resources 226 0 R /Group 273 0 R /MediaBox [ 0 0 612 792 ] /Annots [ 275 0 R 276 0 R 277 0 R ] /Type /Page >> endobj 11 0 obj << /Contents 278 0 R /Parent 1 0 R /Resources 279 0 R /Group 304 0 R /MediaBox [ 0 0 612 792 ] /Annots [ 306 0 R 307 0 R 308 0 R ] /Type /Page >> endobj 12 0 obj << /Contents 309 0 R /Parent 1 0 R /Resources 310 0 R /Group 329 0 R /MediaBox [ 0 0 612 792 ] /Annots [ 331 0 R 332 0 R 333 0 R 334 0 R 335 0 R 336 0 R 337 0 R 338 0 R 339 0 R 340 0 R ] /Type /Page >> endobj 13 0 obj << /Contents 341 0 R /Parent 1 0 R /Type /Page /Resources 342 0 R /MediaBox [ 0 0 612 792 ] >> endobj 14 0 obj << /Contents 343 0 R /Parent 1 0 R /Type /Page /Resources 344 0 R /MediaBox [ 0 0 612 792 ] >> endobj 15 0 obj << /Contents 345 0 R /Parent 1 0 R /Type /Page /Resources 346 0 R /MediaBox [ 0 0 612 792 ] >> endobj 16 0 obj << /Length 3250 /Filter /FlateDecode >> stream xڕYKFϯБqn>*Yq*ofv}Тz$f(RÓ_HJN* | E*Z?ZUea&8̓LZz}HXh\LqJ,̵\VTf[&Qj_O}y,lj\+CFak秦_ەp~M"0Nj~\EaZ=iGaVw@$NnR2$a$Կ: xӖO]G"t\y@%83iLl/|· G$UqDj\c~1Xϴk"Fa*Ӆu] bteB5:>Ue@( O Ja}ZkqM1E%!~c73^GV?jwd=Wu. Y(R^ۋuCZ\X0:̤iȰڿWd"TI>wtib8u?wl'>RXHuV&i(WۮoM; XbҺa$ %,IV*ICfGJfY0rpfzGXhO3ԜʴryL`i3Se7: (ĵ(.Jw.)fEdn4e[imC].\o$ .xHX(,|ө*R~C pXP]yDɚS@&9v44y1Aw4vӜ0hʚ6^+rĬl/=kKYټ| CDxxյ*;S塺 N/靱wdDl+0-e iͮfUq&O~4ԛ'T\xæW=w_)