Inference and Learning for Active Sensing, Experimental Design and Control

Abstract

Abstract. In this paper we argue that maximum expected utility is a suitable framework for modeling a broad range of decision problems arising in pattern recognition and related fields. Examples include, among others, gaze planning and other active vision problems, active learning, sensor and actuator placement and coordination, intelligent humancomputer interfaces, and optimal control. Following this remark, we present a common inference and learning framework for attacking these problems. We demonstrate this approach on three examples: (i) active sensing with nonlinear, non-Gaussian, continuous models, (ii) optimal experimental design to discriminate among competing scientific models, and (iii) nonlinear optimal control. The Principle of Maximum Expected Utility

Bayesian Optimization in a Billion Dimensions via Random Embeddings

“…There also exist several more involved Bayesian non-linear experimental design approaches for constructing the acquisition function, where the utility to be optimized involves an entropy of an aspect of the posterior. This includes the work of Hennig and Schuler (2012) for finding maxima of functions, the works of Kueck, de Freitas, and Doucet (2006) and Kueck, Hoffman, Doucet, and de Freitas (2009) for learning functions, and the work of Hoffman, Kueck, de Freitas, and Doucet (2009) for estimating Markov decision processes. These works rely on expensive approximate inference methods for computing intractable integrals.…”