BoTorch: Programmable Bayesian Optimization in PyTorch
By Maximilian Balandat, Brian Karrer, Daniel R. Jiang, Sam Daulton, Benjamin Letham, Andrew Gordon Wilson, Eytan Bakshy. 2019
102 citations recorded by the source corpus.
Bayesian optimization provides sample-efficient global optimization for a broad range of applications, including automatic machine learning, engineering, physics, and experimental design. We introduce BoTorch, a modern programming framework for Bayesian optimization that combines Monte-Carlo (MC) acquisition functions, a novel sample average approximation optimization approach, auto-differentiation, and variance reduction techniques. BoTorch's modular design facilitates flexible specification and optimization of probabilistic models written in PyTorch, simplifying implementation of new acquisition functions. Our approach is backed by novel theoretical convergence results and made practical by a distinctive algorithmic foundation that leverages fast predictive distributions, hardware acceleration, and deterministic optimization. In experiments, we demonstrate the improved sample efficiency of BoTorch relative to other popular libraries.
Subfields: Artificial Intelligence, Management Science and Operations Research
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Colin White
- Yash Savani
- Willie Neiswanger
- Maximilian Balandat
- Sam Daulton
- D.A. Forsyth
- Eytan Bakshy
- Min Jin Chong
In Management Science and Operations Research
- Maximilian Balandat
- Sam Daulton
- Yash Savani
- D.A. Forsyth
- Colin White
- Willie Neiswanger
- Eytan Bakshy
- Min Jin Chong
Citations in the corpus
- BANANAS: Bayesian Optimization with Neural Architectures for Neural Architecture Search (2019, methodology): “Furthermore, local optimization methods such as mutation, evolution, and local search have been shown to be effective for acquisition function optimization (Balandat et al. 2019; Kandasamy et al. 2018b; Wilson, Hutter, and Deisenroth 2018).”
- Effectively Unbiased FID and Inception Score and Where to Find Them (2019, methodology): “For the implementation of Sobol sequence, we use QMC sampler from BoTorch [2].”
- Differentiable Expected Hypervolume Improvement for Parallel Multi-Objective Bayesian Optimization (2020, methodology): “[5] Maximilian Balandat, Brian Karrer, Daniel R.”