Research map / Federated learning and privacy / Differential privacy
Private Reinforcement Learning with PAC and Regret Guarantees
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add private counters to optimism-based episodic RL | PUCB · Tabular episodic MDP (PAC) | sample complexity: \frac{SAH^4}{\alpha^2} + \frac{S^2AH^4}{\varepsilon\alpha} | Mixed |
| Add private counters to optimism-based episodic RL | PUCB · Tabular episodic MDP (regret) | regret: H^2\sqrt{SAT} + \frac{SAH^3 + S^2AH^3}{\varepsilon} | Mixed |
| Require ε-JDP privacy | Episodic tabular RL · Tabular episodic MDP (PAC) | sample complexity lower bound: \frac{SAH}{\alpha\varepsilon}\ln\left(\frac{1}{\beta} \right) | Hurt |
| Require ε-JDP privacy | Episodic tabular RL · Tabular episodic MDP (regret) | regret lower bound: \frac{S A H\log(T)}{\varepsilon} | Hurt |
| Use privacy-preserving counters whose noise scales with episode length | PUCB · Tabular episodic MDP | privacy cost in utility bounds: E_{\varepsilon} = \frac{3}{\varepsilon}H\log\left(\frac{2SAH+S^2AH}{\beta'}\right)\log\left(T\right)^{5/2} | Helped |
Closest papers by meaning
- Differentially Private Reinforcement Learning with Self-Play (NeurIPS 2024 · similarity 0.92)
- On the Sample Complexity of Differentially Private Policy Optimization (NeurIPS 2025 · similarity 0.91)
- Towards Differentially Private Reinforcement Learning with General Function Approximation (NeurIPS 2026 · similarity 0.90)
- Differentially Private Sparse Reward Estimation with Preference Feedback (NeurIPS 2026 · similarity 0.87)
- Connecting Thompson Sampling and UCB: Towards More Efficient Trade-offs Between Privacy and Regret (ICML 2025 · similarity 0.87)
- Shuffle and Joint Differential Privacy for Generalized Linear Contextual Bandits (NeurIPS 2026 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.