# Deceptive Reinforcement Learning for Privacy-Preserving Planning

arXiv: https://arxiv.org/abs/2102.03022 · HTML: https://mlsanity.com/paper/2102.03022/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 41 references that are in the map.
- First posted on arXiv 5 Feb 2021.

## Builds on (most cited first, 7 of 7)

- [Private Reinforcement Learning with PAC and Regret Guarantees](https://mlsanity.com/paper/2009.09052/) (2020)
- [Deceptive Reinforcement Learning Under Adversarial Manipulations on Cost Signals](https://mlsanity.com/paper/1906.10571/) (2019)
- [Explicability? Legibility? Predictability? Transparency? Privacy? Security? The Emerging Landscape of Interpretable Agent Behavior](https://mlsanity.com/paper/1811.09722/) (2018)
- [Privacy-preserving Q-Learning with Functional Noise in Continuous State Spaces](https://mlsanity.com/paper/1901.10634/) (2019)
- [A Unified Framework for Planning in Adversarial and Cooperative Environments](https://mlsanity.com/paper/1802.06137/) (2018)
- [Heuristic Online Goal Recognition in Continuous Domains](https://mlsanity.com/paper/1709.09839/) (2017)
- [Deception in Optimal Control](https://mlsanity.com/paper/1805.03090/) (2018)

## Built on by (most cited first, 14 of 14)

- [Reinforcement Learning for Feedback-Enabled Cyber Resilience](https://mlsanity.com/paper/2107.00783/) (2021)
- [New Challenges in Reinforcement Learning: A Survey of Security and Privacy](https://mlsanity.com/paper/2301.00188/) (2022)
- [Deceptive Reinforcement Learning in Model-Free Domains](https://mlsanity.com/paper/2303.10838/) (2023)
- [adaPARL: Adaptive Privacy-Aware Reinforcement Learning for Sequential-Decision Making Human-in-the-Loop Systems](https://mlsanity.com/paper/2303.04257/) (2023)
- [Multi-Agent Reinforcement Learning: Methods, Applications, Visionary Prospects, and Challenges](https://mlsanity.com/paper/2305.10091/) (2023)
- [Controlling Federated Learning for Covertness](https://mlsanity.com/paper/2308.08825/) (2023)
- [Covert Planning against Imperfect Observers](https://mlsanity.com/paper/2310.16791/) (2023)
- [Planning for Attacker Entrapment in Adversarial Settings](https://mlsanity.com/paper/2303.00822/) (2023)
- [Deceptive Path Planning via Reinforcement Learning with Graph Neural Networks](https://mlsanity.com/paper/2402.06552/) (2024)
- [Preserving the Privacy of Reward Functions in MDPs through Deception](https://mlsanity.com/paper/2407.09809/) (2024)
- [Value of Information-based Deceptive Path Planning Under Adversarial Interventions](https://mlsanity.com/paper/2503.24284/) (2025)
- [Resilient Strategies for Stochastic Systems: How Much Does It Take to Break a Winning Strategy?](https://mlsanity.com/paper/2602.24191/) (2026)
- [Detection of adversarial intent in Human-AI teams using LLMs](https://mlsanity.com/paper/2603.20976/) (2026)
- [Repeated Deceptive Path Planning against Learnable Observer](https://mlsanity.com/paper/2605.07174/) (2026)

## Abstract (arXiv)

In this paper, we study the problem of deceptive reinforcement learning to preserve the privacy of a reward function. Reinforcement learning is the problem of finding a behaviour policy based on rewards received from exploratory behaviour. A key ingredient in reinforcement learning is a reward function, which determines how much reward (negative or positive) is given and when. However, in some situations, we may want to keep a reward function private; that is, to make it difficult for an observer to determine the reward function used. We define the problem of privacy-preserving reinforcement learning, and present two models for solving it. These models are based on dissimulation -- a form of deception that `hides the truth'. We evaluate our models both computationally and via human behavioural experiments. Results show that the resulting policies are indeed deceptive, and that participants can determine the true reward function less reliably than that of an honest agent.