More Like this

1. Deductive Synthesis of Reinforcement Learning Agents for Infinite Horizon Tasks

   Wang, Yuning; Zhu, He (July 2025, 37th International Conference on Computer Aided Verification (CAV))

   We propose a deductive synthesis framework for construct- ing reinforcement learning (RL) agents that provably satisfy temporal reach-avoid specifications over infinite horizons. Our approach decomposes these temporal specifications into a sequence of finite-horizon subtasks, for which we synthesize individual RL policies. Using formal verification techniques, we ensure that the composition of a finite number of subtask policies guarantees satisfaction of the overall specification over infinite horizons. Experimental results on a suite of benchmarks show that our synthesized agents outperform standard RL methods in both task performance and compliance with safety and temporal requirements. 

   more » « less
2. Deductive Synthesis of Reinforcement Learning Agents for Infinite Horizon Tasks

   https://doi.org/10.1007/978-3-031-98685-7_4  

   Wang, Yuning; Zhu, He (July 2025, Springer Nature Switzerland)

   Abstract We propose a deductive synthesis framework for constructing reinforcement learning (RL) agents that provably satisfy temporal reach-avoid specifications over infinite horizons. Our approach decomposes these temporal specifications into a sequence of finite-horizon subtasks, for which we synthesize individual RL policies. Using formal verification techniques, we ensure that the composition of a finite number of subtask policies guarantees satisfaction of the overall specification over infinite horizons. Experimental results on a suite of benchmarks show that our synthesized agents outperform standard RL methods in both task performance and compliance with safety and temporal requirements. 

   more » « less
3. Instructing Goal-Conditioned Reinforcement Learning Agents with Temporal Logic Objectives

   Qiu, Wenjie; Mao, Wensen; Zhu, He (December 2023, Advances in Neural Information Processing Systems (NeurIPS))

   Goal-conditioned reinforcement learning (RL) is a powerful approach for learning general-purpose skills by reaching diverse goals. However, it has limitations when it comes to task-conditioned policies, where goals are specified by temporally extended instructions written in the Linear Temporal Logic (LTL) formal language. Existing approaches for finding LTL-satisfying policies rely on sampling a large set of LTL instructions during training to adapt to unseen tasks at inference time. However, these approaches do not guarantee generalization to out-of-distribution LTL objectives, which may have increased complexity. In this paper, we propose a novel approach to address this challenge. We show that simple goal-conditioned RL agents can be instructed to follow arbitrary LTL specifications without additional training over the LTL task space. Unlike existing approaches that focus on LTL specifications expressible as regular expressions, our technique is unrestricted and generalizes to ω-regular expressions. Experiment results demonstrate the effectiveness of our approach in adapting goal-conditioned RL agents to satisfy complex temporal logic task specifications zero-shot. 

   more » « less
4. Eventual Discounting Temporal Logic Counterfactual Experience Replay

   Cameron Voloshin; Abhinav Verma; Yisong Yue (January 2023, arXivorg)

   Linear temporal logic (LTL) offers a simplified way of specifying tasks for policy optimization that may otherwise be difficult to describe with scalar reward functions. However, the standard RL framework can be too myopic to find maximally LTL satisfying policies. This paper makes two contributions. First, we develop a new value-function based proxy, using a technique we call eventual discounting, under which one can find policies that satisfy the LTL specification with highest achievable probability. Second, we develop a new experience replay method for generating off-policy data from on-policy rollouts via counterfactual reasoning on different ways of satisfying the LTL specification. Our experiments, conducted in both discrete and continuous state-action spaces, confirm the effectiveness of our counterfactual experience replay approach. 

   more » « less
5. MUTEX: Learning Unified Policies from Multimodal Task Specifications

   Shah, Rutav; Martín-Martín, Roberto; Zhu, Yuke (November 2023, Conference on Robot Learning)

   Humans use different modalities, such as speech, text, images, videos, etc., to communicate their intent and goals with teammates. For robots to become better assistants, we aim to endow them with the ability to follow instructions and understand tasks specified by their human partners. Most robotic policy learning methods have focused on one single modality of task specification while ignoring the rich cross-modal information. We present MUTEX, a unified approach to policy learning from multimodal task specifications. It trains a transformer-based architecture to facilitate cross-modal reasoning, combining masked modeling and cross-modal matching objectives in a two-stage training procedure. After training, MUTEX can follow a task specification in any of the six learned modalities (video demonstrations, goal images, text goal descriptions, text instructions, speech goal descriptions, and speech instructions) or a combination of them. We systematically evaluate the benefits of MUTEX in a newly designed dataset with 100 tasks in simulation and 50 tasks in the real world, annotated with multiple instances of task specifications in different modalities, and observe improved performance over methods trained specifically for any single modality. 

   more » « less