How Private Is Your RL Policy? An Inverse RL Based Analysis Framework
Kritika Prakash, Fiza Husain, Praveen Paruchuri, Sujit Gujar
Abstract
Reinforcement Learning (RL) enables agents to learn how to perform various tasks from scratch. In domains like autonomous driving, recommendation systems and more, optimal RL policies learned could cause a privacy breach if the policies memorize any part of the private reward. We study the set of existing differentially-private RL policies derived from various RL algorithms such as Value Iteration, Deep Q Networks, and Vanilla Proximal Policy Optimization. We propose a new Privacy-Aware Inverse RL (PRIL) analysis framework, that performs reward reconstruction as an adversarial attack on private policies that the agents may deploy. For this, we introduce the reward reconstruction attack, wherein we seek to reconstruct the original reward from a privacy-preserving policy using an Inverse RL algorithm. An adversary must do poorly at reconstructing the original reward function if the agent uses a tightly private policy. Using this framework, we empirically test the effectiveness of the privacy guarantee offered by the private algorithms on multiple instances of the FrozenLake domain of varying complexities. Based on the analysis performed, we infer a gap between the current standard of privacy offered and the standard of privacy needed to protect reward functions in RL. We do so by quantifying the extent to which each private policy protects the reward function by measuring distances between the original and reconstructed rewards. Recent works use Differential Privacy (DP) [7] to make the agent's policy quantifiably private and to get a rigorous privacy guarantee. Like many other areas in AI, RL has also started adopting DP to establish a mathematical way of guaranteeing data privacy in RL environments. However, an important question is: does the privacy guarantee offered by the private policies translate well into protecting the reward function? If not, how can we understand the gap in
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
Builds on4
- Deep Learning with Differential PrivacyMartín Abadi, Andy Chu, Ian J. Goodfellow, H. Brendan McMahan et al.CCS 2016 · 7,620 citations
- Adversarial Policies: Attacking Deep Reinforcement LearningAdam Gleave, Michael Dennis, Cody Wild, Neel Kant et al.ICLR 2020 · 415 citations
- Robust Deep Reinforcement Learning through Adversarial LossTuomas P. Oikarinen, Wang Zhang, Alexandre Megretski, Luca Daniel et al.NeurIPS 2021 · 134 citations
- Private Reinforcement Learning with PAC and Regret GuaranteesGiuseppe Vietri, Borja Balle, Akshay Krishnamurthy, Zhiwei Steven WuICML 2020 · 70 citations
Related papers
- Trust Region Reward Optimization and Proximal Inverse Reward Optimization AlgorithmYang Chen, Menglin Zou, Jiaqi Zhang, Yitan Zhang et al.NeurIPS 2025 · 1 citation
- Reinforcement Learning-Based Black-Box Model Inversion AttacksGyojin Han, Jaehyun Choi, Haeil Lee, Junmo KimCVPR 2023
- Towards the Transferability of Rewards Recovered via Regularized Inverse Reinforcement LearningAndreas Schlaginhaufen, Maryam KamgarpourNeurIPS 2024 · 8 citations
- Provably Efficient Learning of Transferable RewardsAlberto Maria Metelli, Giorgia Ramponi, Alessandro Concetti, Marcello RestelliICML 2021 · 36 citations
- Trust Region Inverse Reinforcement Learning: Explicit Dual Ascent using Local Policy UpdatesAnish Abhijit Diwan, Davide Tateo, Christopher Mower, Haitham Bou Ammar et al.ICML 2026
