Timezone: »
Bayesian optimization usually assumes that a Bayesian prior is given. However, the strong theoretical guarantees in Bayesian optimization are often regrettably compromised in practice because of unknown parameters in the prior. In this paper, we adopt a variant of empirical Bayes and show that, by estimating the Gaussian process prior from offline data sampled from the same prior and constructing unbiased estimators of the posterior, variants of both GP-UCB and \emph{probability of improvement} achieve a near-zero regret bound, which decreases to a constant proportional to the observational noise as the number of offline data and the number of online evaluations increase. Empirically, we have verified our approach on challenging simulated robotic problems featuring task and motion planning.
Author Information
Zi Wang (MIT)
Beomjoon Kim (MIT)
Leslie Kaelbling (MIT)
Related Events (a corresponding poster, oral, or spotlight)
-
2018 Poster: Regret bounds for meta Bayesian optimization with an unknown Gaussian process prior »
Wed Dec 5th through Thu the 6th Room Room 210
More from the Same Authors
-
2021 Poster: Tailoring: encoding inductive biases by optimizing unsupervised objectives at prediction time »
Ferran Alet · Maria Bauza · Kenji Kawaguchi · Nurullah Giray Kuru · Tomás Lozano-Pérez · Leslie Kaelbling -
2021 Poster: Understanding End-to-End Model-Based Reinforcement Learning Methods as Implicit Parameterization »
Clement Gehring · Kenji Kawaguchi · Leslie Kaelbling -
2020 Poster: Adversarially-learned Inference via an Ensemble of Discrete Undirected Graphical Models »
Adarsh Keshav Jeewajee · Leslie Kaelbling -
2019 Poster: Neural Relational Inference with Fast Modular Meta-learning »
Ferran Alet · Erica Weng · Tomás Lozano-Pérez · Leslie Kaelbling -
2018 Workshop: Infer to Control: Probabilistic Reinforcement Learning and Structured Control »
Leslie Kaelbling · Martin Riedmiller · Marc Toussaint · Igor Mordatch · Roy Fox · Tuomas Haarnoja -
2015 Poster: Bayesian Optimization with Exponential Convergence »
Kenji Kawaguchi · Leslie Kaelbling · Tomás Lozano-Pérez -
2008 Poster: Multi-Agent Filtering with Infinitely Nested Beliefs »
Luke Zettlemoyer · Brian Milch · Leslie Kaelbling -
2008 Spotlight: Multi-Agent Filtering with Infinitely Nested Beliefs »
Luke Zettlemoyer · Brian Milch · Leslie Kaelbling -
2007 Workshop: The Grammar of Vision: Probabilistic Grammar-Based Models for Visual Scene Understanding and Object Categorization »
Virginia Savova · Josh Tenenbaum · Leslie Kaelbling · Alan L Yuille