Bayesian Optimal Control of Smoothly Parameterized Systems

By Yasin Abbasi-Yadkori, Csaba Szepesvari. 2015

49 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Efficient Exploration Through Bayesian Deep Q-Networks (2018, methodology): “[3] Abbasi-Yadkori, Y.”
Model-Based Reinforcement Learning with Value-Targeted Regression (2020, methodology): “Our model (1) provides a linearized parameterized model, which can be used to approximate more general smooth parametric robotic systems Abbasi-Yadkori and Szepesvári (2015). For another example, consider large-scale queueing networks where the arrival rate and job processing speed for each queue is”
Data center cooling using model-predictive control (2018, background): “Similar regret bounds can be derived using Thompson sampling in place of optimization [3, 4, 25], but most of these approaches make unrealistic stability assumptions about the intermediate controllers, and can in practice induce diverging state trajectories in early stages.”
Regret Minimization for Reinforcement Learning by Evaluating the Optimal Bias Function (2019, background): “Because of the simplicity and scalability, as well as the provably optimal regret bound, the posterior sampling has been getting popular in the related research ﬁeld [Osband et al., 2013, Osband and Van Roy, 2016, Agrawal and Jia, 2017, Abbasi-Yadkori, 2015].”
Convex Optimization: Algorithms and Complexity (2015, unknown): “4) and its analysis were also extended to the continuous state and action space and average-cost setting, under the assumption of smoothly parameterizable dynamics [Abbasi-Yadkori and Szepesvari, 2015]. The main algorithmic modification in this extension is to incorporate a specific schedule for upd”