Sample-Efficient Reinforcement Learning with Temporal Logic Objectives: Leveraging the Task Specification to Guide Exploration

2024-10-16 · Yiannis Kantaros, Jun Wang

This paper addresses the problem of learning optimal control policies for systems with uncertain dynamics and high-level control objectives specified as Linear Temporal Logic (LTL) formulas. Uncertainty is considered in the workspace structure and the outcomes of control decisions giving rise to an unknown Markov Decision Process (MDP). Existing reinforcement learning (RL) algorithms for LTL tasks typically rely on exploring a product MDP state-space uniformly (using e.g., an $\epsilon$-greedy policy) compromising sample-efficiency. This issue becomes more pronounced as the rewards get sparser and the MDP size or the task complexity increase. In this paper, we propose an accelerated RL algorithm that can learn control policies significantly faster than competitive approaches. Its sample-efficiency relies on a novel task-driven exploration strategy that biases exploration towards directions that may contribute to task satisfaction. We provide theoretical analysis and extensive comparative experiments demonstrating the sample-efficiency of the proposed method. The benefit of our method becomes more evident as the task complexity or the MDP size increases.

📄 PDF Abstract BibTeX arXiv:2410.12136

Code (1)

kantaroslab/accrl 공식 구현 pytorch

Tasks

Reinforcement Learning (RL)

Accelerated Reinforcement Learning for Temporal Logic Control Objectives

2022-05-09 · Yiannis Kantaros

This paper addresses the problem of learning control policies for mobile robots, modeled as unknown Markov Decision Processes (MDPs), that are tasked with temporal logic missions, such as sequencing, coverage, or surveil…

Model-based Reinforcement Learning reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Computably Continuous Reinforcement-Learning Objectives are PAC-learnable

2023-03-09 · Cambridge Yang, Michael Littman, Michael Carbin

In reinforcement learning, the classic objectives of maximizing discounted and finite-horizon cumulative rewards are PAC-learnable: There are algorithms that learn a near-optimal policy with high probability using a fini…

General Reinforcement Learning reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

A PAC Learning Algorithm for LTL and Omega-regular Objectives in MDPs

2023-10-18 · Mateo Perez, Fabio Somenzi, Ashutosh Trivedi

Linear temporal logic (LTL) and omega-regular objectives -- a superset of LTL -- have seen recent use as a way to express non-Markovian objectives in reinforcement learning. We introduce a model-based probably approximat…

PAC learning reinforcement-learning

Reinforcement Learning with General LTL Objectives is Intractable

2021-11-21 · AAAI Workshop CLeaR 2022 2 · Cambridge Yang, Michael Littman, Michael Carbin

In recent years, researchers have made significant progress in devising reinforcement-learning algorithms for optimizing linear temporal logic (LTL) objectives and LTL-like objectives. Despite these advancements, there a…

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

On the (In)Tractability of Reinforcement Learning for LTL Objectives

2021-11-24 · Cambridge Yang, Michael Littman, Michael Carbin