A Survey on the Verification of Reinforcement Learning Policies

Luca Marzari; Ezio Bartocci; Enrico Marchesini

IJCAI 20260 citations

A Survey on the Verification of Reinforcement Learning Policies

Luca Marzari, Ezio Bartocci, Enrico Marchesini

Abstract

Reinforcement learning (RL) is increasingly applied in complex, safety-critical domains, yet the lack of rigorous behavioral guarantees for neural network-based policies remains a major barrier to deployment. Recent advances in policy expressiveness and scale have intensified this challenge, leading to a rapidly growing but conceptually fragmented body of work on RL policy verification. This survey provides a unifying perspective on RL verification methods. We introduce a taxonomy that clarifies relationships among existing approaches along three axes: verification paradigm (formal versus probabilistic), temporal scope (step-wise versus multi-step), and guarantees strength. Beyond taxonomy, we unify underlying theoretical foundations, make implicit assumptions and limitations explicit, and identify emerging directions.

Agent-based and Multi-agent Systems: Formal verification, validation and synthesisMachine Learning: Reinforcement learning

BibTeX

@inproceedings{ijcai2026_asurveyontheveri,
  title = {A Survey on the Verification of Reinforcement Learning Policies},
  author = {Luca Marzari and Ezio Bartocci and Enrico Marchesini},
  booktitle = {IJCAI 2026},
  year = {2026}
}