Accelerated Bayesian Optimization throughWeight-Prior Tuning
This work addresses the challenge of improving BO efficiency in scenarios where auxiliary data is available but not directly usable, offering a domain-specific solution for optimization tasks like manufacturing.
The paper tackles the problem of accelerating Bayesian optimization (BO) for expensive-to-evaluate functions by leveraging auxiliary data with similar covariance structure to construct a more appropriate weight prior for the Gaussian Process model, resulting in demonstrated acceleration on test functions and a practical application in short-polymer fibre manufacture.
Bayesian optimization (BO) is a widely-used method for optimizing expensive (to evaluate) problems. At the core of most BO methods is the modeling of the objective function using a Gaussian Process (GP) whose covariance is selected from a set of standard covariance functions. From a weight-space view, this models the objective as a linear function in a feature space implied by the given covariance K, with an arbitrary Gaussian weight prior ${\bf w} \sim \mathcal{N} ({\bf 0}, {\bf I})$. In many practical applications there is data available that has a similar (covariance) structure to the objective, but which, having different form, cannot be used directly in standard transfer learning. In this paper we show how such auxiliary data may be used to construct a GP covariance corresponding to a more appropriate weight prior for the objective function. Building on this, we show that we may accelerate BO by modeling the objective function using this (learned) weight prior, which we demonstrate on both test functions and a practical application to short-polymer fibre manufacture.