9 papers
Jaxolotl: A Unified High-Performance Benchmark Suite for LTL-Based Multi-Task RL
Mathias Jackermeier, Jacques Cloete, Alessandro Abate
Training agents to follow arbitrary instructions is an important goal of multi-task reinforcement learning (RL). Linear temporal logic (LTL) provides a precise and structured forma…
Efficient Bayes-Adaptive Reinforcement Learning with Temporal Logic Specifications
Jonathan Hau, Alessandro Abate
We present a novel end-to-end model-based Reinforcement Learning (RL) algorithm for efficient policy synthesis under given Linear Temporal Logic (LTL) specifications (e.g., safety…
Robust Parameter Learning for Uncertain MDPs
Yannik Schnitzer, Alessandro Abate, David Parker
Learning-based approaches to verifying unknown Markov decision processes (MDPs) often employ uncertain MDPs. These models use, for example, confidence intervals to capture transiti…
Zero-Shot Instruction Following in RL via Structured LTL Representations
Mathias Jackermeier, Mattia Giuri, Jacques Cloete +1
We study instruction following in multi-task reinforcement learning, where an agent must zero-shot execute novel tasks not seen during training. In this setting, linear temporal lo…
Semantically Labelled Automata for Multi-Task Reinforcement Learning with LTL Instructions
Alessandro Abate, Giuseppe De Giacomo, Mathias Jackermeier +3
We study multi-task reinforcement learning (RL), a setting in which an agent learns a single, universal policy capable of generalising to arbitrary, possibly unseen tasks. We consi…
Probabilistic Performance Guarantees for Multi-Task Reinforcement Learning
Yannik Schnitzer, Mathias Jackermeier, Alessandro Abate +1
Multi-task reinforcement learning trains generalist policies that can execute multiple tasks. While recent years have seen significant progress, existing approaches rarely provide…
PlatoLTL: Scaling LTL-Guided Multi-Task RL
Jacques Cloete, Mathias Jackermeier, Ioannis Havoutis +1
Linear temporal logic (LTL) has emerged as a powerful formalism for specifying structured, temporally extended tasks in multi-task reinforcement learning (RL). However, while exist…
Zero-Shot Instruction Following in RL via Structured LTL Representations
Mattia Giuri, Mathias Jackermeier, Alessandro Abate
Linear temporal logic (LTL) is a compelling framework for specifying complex, structured tasks for reinforcement learning (RL) agents. Recent work has shown that interpreting LTL i…
DeepLTL: Learning to Efficiently Satisfy Complex LTL Specifications for Multi-Task RL
Mathias Jackermeier, Alessandro Abate
Linear temporal logic (LTL) has recently been adopted as a powerful formalism for specifying complex, temporally extended tasks in multi-task reinforcement learning (RL). However,…