Timezone: »
Reinforcement learning (RL) aims to find an optimal policy by interaction with an environment. Consequently, learning complex behavior requires a vast number of samples, which can be prohibitive in practice. Nevertheless, instead of systematically reasoning and actively choosing informative samples, policy gradients for local search are often obtained from random perturbations. These random samples yield high variance estimates and hence are sub-optimal in terms of sample complexity. Actively selecting informative samples is at the core of Bayesian optimization, which constructs a probabilistic surrogate of the objective from past samples to reason about informative subsequent ones. In this paper, we propose to join both worlds. We develop an algorithm utilizing a probabilistic model of the objective function and its gradient. Based on the model, the algorithm decides where to query a noisy zeroth-order oracle to improve the gradient estimates. The resulting algorithm is a novel type of policy search method, which we compare to existing black-box algorithms. The comparison reveals improved sample complexity and reduced variance in extensive empirical evaluations on synthetic objectives. Further, we highlight the benefits of active sampling on popular RL benchmarks.
Author Information
Sarah Müller (University of Tübingen)
Alexander von Rohr (RWTH Aachen University)
Sebastian Trimpe (RWTH Aachen University)
More from the Same Authors
-
2021 : Towards Safe Global Optimality in Robot Learning with GoSafe »
Bhavya Sukhija · Matteo Turchetta · Andreas Krause · Sebastian Trimpe · Dominik Baumann -
2023 Poster: On kernel-based statistical learning theory in the mean field limit »
Christian Fiedler · Michael Herty · Sebastian Trimpe -
2021 : Learning By Doing: Controlling a Dynamical System using Control Theory, Reinforcement Learning, or Causality + Q&A »
Sebastian Weichwald · Niklas Pfister · Dominik Baumann · Isabelle Guyon · Oliver Kroemer · Tabitha Lee · Søren Wengel Mogensen · Jonas Peters · Sebastian Trimpe