A Nonparametric Conjugate Prior Distribution for the Maximizing Argument of a Noisy Function
This addresses stochastic optimization problems for researchers and practitioners, offering a novel method but appears incremental as it builds on prior Bayesian approaches.
The paper tackles the problem of finding extrema of noisy, nonlinear functions by proposing a Bayesian approach that directly models the distribution over extrema, skipping explicit function representation, and demonstrates effectiveness on a noisy, high-dimensional, non-convex objective function.
We propose a novel Bayesian approach to solve stochastic optimization problems that involve finding extrema of noisy, nonlinear functions. Previous work has focused on representing possible functions explicitly, which leads to a two-step procedure of first, doing inference over the function space and second, finding the extrema of these functions. Here we skip the representation step and directly model the distribution over extrema. To this end, we devise a non-parametric conjugate prior based on a kernel regressor. The resulting posterior distribution directly captures the uncertainty over the maximum of the unknown function. We illustrate the effectiveness of our model by optimizing a noisy, high-dimensional, non-convex objective function.