High Dimensional Bayesian Optimization Using Dropout
This addresses the problem of high-dimensional optimization for researchers and practitioners in machine learning and materials science, but it appears incremental as it builds on existing methods with a novel dropout approach.
The paper tackles the challenge of scaling Bayesian optimization to high dimensions by proposing a dropout strategy that optimizes only a subset of variables per iteration, demonstrating efficacy on benchmark functions and real-world applications like training cascade classifiers and optimizing alloy composition.
Scaling Bayesian optimization to high dimensions is challenging task as the global optimization of high-dimensional acquisition function can be expensive and often infeasible. Existing methods depend either on limited active variables or the additive form of the objective function. We propose a new method for high-dimensional Bayesian optimization, that uses a dropout strategy to optimize only a subset of variables at each iteration. We derive theoretical bounds for the regret and show how it can inform the derivation of our algorithm. We demonstrate the efficacy of our algorithms for optimization on two benchmark functions and two real-world applications- training cascade classifiers and optimizing alloy composition.