Gaussian Process Optimization with Mutual Information
arXiv:1311.4825
Abstract
In this paper, we analyze a generic algorithm scheme for sequential global optimization using Gaussian processes. The upper bounds we derive on the cumulative regret for this generic algorithm improve by an exponential factor the previously known bounds for algorithms like GP-UCB. We also introduce the novel Gaussian Process Mutual Information algorithm (GP-MI), which significantly improves further these upper bounds for the cumulative regret. We confirm the efficiency of this algorithm on synthetic and real tasks against the natural competitor, GP-UCB, and also the Expected Improvement heuristic.
Proceedings of The 31st International Conference on Machine Learning (ICML 2014)
References in corpus (5)
- Practical Bayesian Optimization of Machine Learning Algorithms
- Exponential inequalities for self-normalized martingales with applications
- Parallelizing Exploration-Exploitation Tradeoffs with Gaussian Process Bandit Optimization
- Exponential Regret Bounds for Gaussian Process Bandits with Deterministic Observations
- Can Small Islands Protect Nearby Coasts From Tsunamis? An Active Experimental Design Approach
Cited by in corpus (13)
- Emulation of Higher-Order Tensors in Manifold Monte Carlo Methods for Bayesian Inverse Problems
- A multifidelity Bayesian optimization method for inertial confinement fusion design
- Efficient batch-sequential Bayesian optimization with moments of truncated Gaussian vectors
- Time-varying Gaussian Process Bandit Optimization with Non-constant Evaluation Time
- Using Distance Correlation for Efficient Bayesian Optimization
- A Simple Heuristic for Bayesian Optimization with A Low Budget
- End-to-End Learning of Deep Kernel Acquisition Functions for Bayesian Optimization
- Statistical Learning and Estimation of Piano Fingering
- Optimization for Gaussian Processes via Chaining
- Simple and Scalable Parallelized Bayesian Optimization
- Towards Bursting Filter Bubble via Contextual Risks and Uncertainties
- Learning Arbitrary Quantities of Interest from Expensive Black-Box Functions through Bayesian Sequential Optimal Design
- Harnessing Heterogeneity: Learning from Decomposed Feedback in Bayesian Modeling