Semantically Labelled Automata for Multi-Task Reinforcement Learning with LTL Instructions

Abate, Alessandro; De Giacomo, Giuseppe; Jackermeier, Mathias; Křetínský, Jan; Prokop, Maximilian; Weinhuber, Christoph

doi:10.24963/ijcai.2026/1

Semantically Labelled Automata for Multi-Task Reinforcement Learning with LTL Instructions

Alessandro Abate, Giuseppe De Giacomo, Mathias Jackermeier, Jan Křetínský, Maximilian Prokop, Christoph Weinhuber

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Main Track. Pages 3-11.

https://doi.org/10.24963/ijcai.2026/1

Keywords

Knowledge Representation and Reasoning: Reasoning about actions
Knowledge Representation and Reasoning: Learning and reasoning
Machine Learning: Multi-task and transfer learning
Agent-based and Multi-agent Systems: Formal verification, validation and synthesis
Machine Learning: Reinforcement learning

Abstract

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 consider tasks specified as linear temporal logic (LTL) formulae, which are commonly used in formal methods to specify properties of systems, and have recently been successfully adopted in RL. In this setting, we present a novel task embedding technique leveraging a new generation of semantic LTL-to-automata translations, originally developed for temporal synthesis. The resulting semantically labelled automata contain rich, structured information in each state that allow us to (i) compute the automaton efficiently on-the-fly, (ii) extract expressive task embeddings used to condition the policy, and (iii) naturally support full LTL. Experimental results in a variety of domains demonstrate that our approach achieves state-of-the-art performance and is able to scale to complex specifications where existing methods fail.

Alessandro Abate, Giuseppe De Giacomo, Mathias Jackermeier, Jan Křetínský, Maximilian Prokop, and Christoph Weinhuber. 2026. Semantically Labelled Automata for Multi-Task Reinforcement Learning with LTL Instructions. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 3–11. https://doi.org/10.24963/ijcai.2026/1

@inproceedings{ijcai2026p1,
  title     = {Semantically Labelled Automata for Multi-Task Reinforcement Learning with LTL Instructions},
  author    = {Abate, Alessandro and De Giacomo, Giuseppe and Jackermeier, Mathias and Křetínský, Jan and Prokop, Maximilian and Weinhuber, Christoph},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {3--11},
  year      = {2026},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2026/1},
  url       = {https://doi.org/10.24963/ijcai.2026/1},
}

TY  - CPAPER
TI  - Semantically Labelled Automata for Multi-Task Reinforcement Learning with LTL Instructions
AU  - Abate, Alessandro
AU  - De Giacomo, Giuseppe
AU  - Jackermeier, Mathias
AU  - Křetínský, Jan
AU  - Prokop, Maximilian
AU  - Weinhuber, Christoph
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 3
EP  - 11
DO  - 10.24963/ijcai.2026/1
UR  - https://doi.org/10.24963/ijcai.2026/1
ER  -