Bayesian Optimization over Discrete and Mixed Spaces via Probabilistic Reparameterization

October 18, 2022 ยท Entered Twilight ยท ๐Ÿ› Neural Information Processing Systems

๐Ÿ’ค TWILIGHT: Eternal Rest
Repo abandoned since publication

Repo contents: .gitignore, BO_Probabilistic_Reparameterization.pdf, CODEOWNERS, CODE_OF_CONDUCT.md, CONTRIBUTING.md, LICENSE, README.md, discrete_mixed_bo, experiments, setup.py

Authors Samuel Daulton, Xingchen Wan, David Eriksson, Maximilian Balandat, Michael A. Osborne, Eytan Bakshy arXiv ID 2210.10199 Category cs.LG: Machine Learning Cross-listed cs.AI, math.OC, stat.ML Citations 57 Venue Neural Information Processing Systems Repository https://github.com/facebookresearch/bo_pr โญ 76 Last Checked 1 month ago
Abstract
Optimizing expensive-to-evaluate black-box functions of discrete (and potentially continuous) design parameters is a ubiquitous problem in scientific and engineering applications. Bayesian optimization (BO) is a popular, sample-efficient method that leverages a probabilistic surrogate model and an acquisition function (AF) to select promising designs to evaluate. However, maximizing the AF over mixed or high-cardinality discrete search spaces is challenging standard gradient-based methods cannot be used directly or evaluating the AF at every point in the search space would be computationally prohibitive. To address this issue, we propose using probabilistic reparameterization (PR). Instead of directly optimizing the AF over the search space containing discrete parameters, we instead maximize the expectation of the AF over a probability distribution defined by continuous parameters. We prove that under suitable reparameterizations, the BO policy that maximizes the probabilistic objective is the same as that which maximizes the AF, and therefore, PR enjoys the same regret bounds as the original BO policy using the underlying AF. Moreover, our approach provably converges to a stationary point of the probabilistic objective under gradient ascent using scalable, unbiased estimators of both the probabilistic objective and its gradient. Therefore, as the number of starting points and gradient steps increase, our approach will recover of a maximizer of the AF (an often-neglected requisite for commonly used BO regret bounds). We validate our approach empirically and demonstrate state-of-the-art optimization performance on a wide range of real-world applications. PR is complementary to (and benefits) recent work and naturally generalizes to settings with multiple objectives and black-box constraints.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

๐Ÿ“œ Similar Papers

In the same crypt โ€” Machine Learning