Receding Horizon Curiosity

TL;DR

This paper introduces a trajectory-optimization-based exploration algorithm for unknown MDPs that improves sample efficiency and model fidelity by directed exploration, outperforming intrinsic motivation methods and maintaining computational efficiency.

Contribution

It presents a novel approximate solution for optimal exploration in unknown MDPs using Bayesian experimental design and trajectory optimization, without prior knowledge of the environment.

Findings

Faster convergence and higher model fidelity compared to intrinsic motivation algorithms.

Maintains computational efficiency over recent model-based active exploration methods.

Effective directed exploration improves sample efficiency in unknown MDPs.

Abstract

Sample-efficient exploration is crucial not only for discovering rewarding experiences but also for adapting to environment changes in a task-agnostic fashion. A principled treatment of the problem of optimal input synthesis for system identification is provided within the framework of sequential Bayesian experimental design. In this paper, we present an effective trajectory-optimization-based approximate solution of this otherwise intractable problem that models optimal exploration in an unknown Markov decision process (MDP). By interleaving episodic exploration with Bayesian nonlinear system identification, our algorithm takes advantage of the inductive bias to explore in a directed manner, without assuming prior knowledge of the MDP. Empirical evaluations indicate a clear advantage of the proposed algorithm in terms of the rate of convergence and the final model fidelity when compared to intrinsic-motivation-based algorithms employing exploration bonuses such as prediction error and information gain. Moreover, our method maintains a computational advantage over a recent model-based active exploration (MAX) algorithm, by focusing on the information gain along trajectories instead of seeking a global exploration policy. A reference implementation of our algorithm and the conducted experiments is publicly available.