Max-value Entropy Search for Efficient Bayesian Optimization
By Zi Wang, Stefanie Sabrina Jegelka. International Conference on Machine Learning, 2017
483 citations recorded by the source corpus.
Entropy Search (ES) and Predictive Entropy Search (PES) are popular and empirically successful Bayesian Optimization techniques. Both rely on a compelling information-theoretic motivation, and maximize the information gained about the $\arg\max$ of the unknown function; yet, both are plagued by the expensive computation for estimating entropies. We propose a new criterion, Max-value Entropy Search (MES), that instead uses the information about the maximum function value. We show relations of MES to other Bayesian optimization methods, and establish a regret bound. We observe that MES maintains or improves the good empirical performance of ES/PES, while tremendously lightening the computational burden. In particular, MES is much more robust to the number of samples used for computing the entropy, and hence more efficient for higher dimensional problems.
Subfields: Artificial Intelligence, Management Science and Operations Research
View Paper: https://www.semanticscholar.org/paper/0d03c970454f55fc3627c340e92a1de55220a304
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
- Guodong Zhang
- Roger Baker Grosse
- Yaochu Jin
- Andrew Gordon Wilson
- Sergey Levine
- Maximilian Balandat
- Sam Daulton
- Eytan Bakshy
In Management Science and Operations Research
- Maximilian Balandat
- Sam Daulton
- Eytan Bakshy
- Yaochu Jin
- Markus Olhofer
- Zhenguo Li
- Zhihua Zhang
- Yujun Li
Citations in the corpus
- An Asymptotically Optimal Multi-Armed Bandit Algorithm and Hyperparameter Optimization (2020, methodology): “…et al., 2011; Bergstra et al., 2011; Springenberg et al., 2016; Srinivas et al., 2010; Hennig and Schuler, 2012; Hernández-Lobato et al., 2014; Wang and Jegelka, 2017; Ru et al., 2018), and the evaluation method (Thornton et al., 2013; Karnin et al., 2013; Jamieson and Talwalkar, 2016; Li et…”
- Recent Advances in Bayesian Optimization (2022, background): “In [287], unknown functions with varying fi-delities are jointly modeled as a convolved Gaussian process [5], then a multi-output random feature approximation is introduced to calculate PES.”
- Functional Variational Bayesian Neural Networks (2019, methodology): “We compared BBB, RBF Random Feature (Rahimi & Recht, 2008) and our fBNNs in the context of Max-value Entropy Search (MES) (Wang & Jegelka, 2017), which requires explicit function samples for Bayesian Optimization.”
- Maximizing acquisition functions for Bayesian optimization (2018, background): “In some cases, this belief is expressed as an explicit ensemble of sample functions [28, 54, 60].”
- BoTorch: Programmable Bayesian Optimization in PyTorch (2019, methodology): “Popular look-ahead acquisition functions include the Knowledge Gradient (KG) (Frazier et al., 2008), Predictive Entropy Search (Henrández-Lobato et al., 2014), Max-Value Entropy Search (Wang & Jegelka, 2017), as well as various heuristics (González et al., 2016b).”