GLASSES: Relieving The Myopia Of Bayesian Optimisation

We present GLASSES: Global optimisation with Look-Ahead through Stochastic Simulation and Expected-loss Search. The majority of global optimisation approaches in use are myopic, in only considering the impact of the next function value; the non-myopic approaches that do exist are able to consider only a handful of future evaluations. Our novel algorithm, GLASSES, permits the consideration of dozens of evaluations into the future. This is done by approximating the ideal look-ahead loss function, which is expensive to evaluate, by a cheaper alternative in which the future steps of the algorithm are simulated beforehand. An Expectation Propagation algorithm is used to compute the expected value of the loss.We show that the far-horizon planning thus enabled leads to substantive performance gains in empirical tests.

[1]  Donald R. Jones,et al.  A Taxonomy of Global Optimization Methods Based on Response Surfaces , 2001, J. Glob. Optim..

[2]  Roman Garnett,et al.  Bayesian optimization for sensor set selection , 2010, IPSN '10.

[3]  Nando de Freitas,et al.  A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning , 2010, ArXiv.

[4]  Ali Jalali,et al.  Hybrid Batch Bayesian Optimization , 2012, ICML.

[5]  Jasper Snoek,et al.  Practical Bayesian Optimization of Machine Learning Algorithms , 2012, NIPS.

[6]  J. Nocedal.  Updating Quasi-Newton Matrices With Limited Storage , 1980 .

[7]  Simon Streltsov,et al.  A Non-myopic Utility Function for Statistical Global Optimization Algorithms , 1999, J. Glob. Optim..

[8]  Ben Taskar,et al.  Near-Optimal MAP Inference for Determinantal Point Processes , 2012, NIPS.

[9]  Tom Minka,et al.  Expectation Propagation for approximate Bayesian inference , 2001, UAI.

[10]  Michael A. Osborne.  Bayesian Gaussian processes for sequential prediction, optimisation and quadrature , 2010 .

[11]  N. Zheng,et al.  Global Optimization of Stochastic Black-Box Systems via Sequential Kriging Meta-Models , 2006, J. Glob. Optim..

[12]  Ben Taskar,et al.  Determinantal Point Processes for Machine Learning , 2012, Found. Trends Mach. Learn..

[13]  Nando de Freitas,et al.  A Bayesian exploration-exploitation approach for optimal online sensing and planning with a visually guided mobile robot , 2009, Auton. Robots.

[14]  Ben Taskar,et al.  Approximate Inference in Continuous Determinantal Processes , 2013, NIPS.

[15]  Michael A. Osborne,et al.  Gaussian Processes for Global Optimization , 2008 .

[16]  Ben Taskar,et al.  k-DPPs: Fixed-Size Determinantal Point Processes , 2011, ICML.

[17]  C. D. Perttunen,et al.  Lipschitzian optimization without the Lipschitz constant , 1993 .

[18]  Scott Sanner,et al.  Sequential Bayesian Optimisation for Spatial-Temporal Monitoring , 2014, UAI.

[19]  Neil D. Lawrence,et al.  Bayesian Optimization for Synthetic Gene Design , 2015, 1505.01627.

[20]  D. Lizotte.  Practical bayesian optimization , 2008 .

[21]  J. Cunningham,et al.  Gaussian Probabilities and Expectation Propagation , 2011, 1111.6832.

[22]  Neil D. Lawrence,et al.  Batch Bayesian Optimization via Local Penalization , 2015, AISTATS.

[23]  D. Ginsbourger,et al.  A Multi-points Criterion for Deterministic Parallel Global Optimization based on Gaussian Processes , 2008 .

[24]  Christopher K. I. Williams,et al.  Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning) , 2005 .