LTL2Action: Generalizing LTL Instructions for Multi-Task RL

Abstract

We address the problem of teaching a deep reinforcement learning (RL) agent to follow instructions in multi-task environments. Instructions are expressed in a well-known formal language -- linear temporal logic (LTL) -- and can specify a diversity of complex, temporally extended behaviours, including conditionals and alternative realizations. Our proposed learning approach exploits the compositional syntax and the semantics of LTL, enabling our RL agent to learn task-conditioned policies that generalize to new instructions, not observed during training. To reduce the overhead of learning LTL semantics, we introduce an environment-agnostic LTL pretraining scheme which improves sample-efficiency in downstream environments. Experiments on discrete and continuous domains target combinatorial task sets of up to $\sim 10^{39}$ unique tasks and demonstrate the strength of our approach in learning to solve (unseen) tasks, given LTL instructions.

International Conference on Machine Learning (ICML) 2021

References in corpus (7)

• Reward Machines: Exploiting Reward Function Structure in Reinforcement Learning
• Benchmarking Batch Deep Reinforcement Learning Algorithms
• Grounded Language Learning in a Simulated 3D World
• Zero-Shot Task Generalization with Multi-Task Deep Reinforcement Learning
• Gated-Attention Architectures for Task-Oriented Language Grounding
• Grounded Language Learning Fast and Slow
• Encoding formulas as deep networks: Reinforcement learning for zero-shot execution of LTL formulas

Cited by in corpus (2)

• Reward Machines: Exploiting Reward Function Structure in Reinforcement Learning
• Compositional Reinforcement Learning from Logical Specifications