On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,657 papers · 148 categories

Trend · papers per month

count % of corpus

48 results for “private reinforcement learning”

Private Reinforcement Learning with PAC and Regret Guaranteescs.LG

Private RL algorithm with privacy guarantees for personalized medicine decisions.

problem Privacy-preserving reinforcement learning for personalized medicine decisions.

method Developed a private optimism-based RL algorithm using joint differential privacy (JDP).

result Achieved strong PAC and regret bounds with a privacy guarantee.

private reinforcement learning personalized medicine joint differential privacy PAC bounds regret bounds

2020-09-18abs ↗pdf ↗

Differentially Private Deep Model-Based Reinforcement Learningcs.LG

PriMORL trains private RL policies on offline data.

problem Private reinforcement learning on offline data.

method PriMORL learns DP models of the environment and optimizes a policy on the penalized private model.

result PriMORL enables training of private RL agents on complex tasks.

differential privacy deep reinforcement learning model-based RL private RL offline RL

2024-02-08abs ↗pdf ↗

Locally Private Distributed Reinforcement Learningcs.LG

We study locally differentially private algorithms for reinforcement learning to obtain a robust policy that performs well across distributed private environments. Our algorithm protects the information of local agents' models from being exploited by adversarial reverse engineering. Since a local policy is strongly bei…

2020-01-31abs ↗pdf ↗

Privacy-Preserving Reinforcement Learning from Human Feedback via Decoupled Reward Modelingstat.ML

Privacy-preserving reinforcement learning from human feedback using decoupled reward modeling.

problem Training large language models with sensitive user information while preserving privacy.

method Proposes a privacy-preserving framework that imposes differential privacy on reward learning only.

result Privacy contributes an additional additive term to the suboptimality gap, and the upper bound is rate-optimal up to logarithmic factors.

privacy-preserving reinforcement learning human feedback differential privacy reward modeling

2026-03-23abs ↗pdf ↗

Differentially Private Reinforcement Learning with Self-Playcs.LG

Study differential privacy in multi-agent RL, achieving efficient and private learning.

problem Protecting sensitive data in multi-agent reinforcement learning.

method Extending DP definitions to two-player games, designing an efficient algorithm with privatized bonuses.

result Achieved trajectory-wise differential privacy in multi-agent RL, improving regret bounds.

Differential Privacy Reinforcement Learning Multi-Agent Systems Privacy Protection

2024-04-11abs ↗pdf ↗

Near-Optimal Differentially Private Reinforcement Learningcs.LG

New algorithm for personalized healthcare with privacy guarantees.

problem Online exploration in reinforcement learning with differential privacy constraints.

method ε-JDP algorithm with privately released exploration bonuses and visitation statistics.

result Regret bound of

O(\sqrt{SAH^2T}+S^2AH^3/ε)

matching information-theoretic lower bound.

Reinforcement Learning Differential Privacy Personalized Healthcare Privacy for Free Regret Bound

2022-12-09abs ↗pdf ↗

Differentially Private Policy Evaluationcs.LG

We present the first differentially private algorithms for reinforcement learning, which apply to the task of evaluating a fixed policy. We establish two approaches for achieving differential privacy, provide a theoretical analysis of the privacy and utility of the two algorithms, and show promising results on simple e…

2016-03-07abs ↗pdf ↗

Towards Optimal Differentially Private Regret Bounds in Linear MDPscs.LG

Improved privacy in RL with near-optimal regret bounds.

problem Privacy-preserving reinforcement learning in personalized decision-making systems.

method Differentially private algorithm based on LSVI-UCB++ with privacy-preserving techniques.

result Achieved a near-optimal regret bound of O(d * sqrt(H^3 * K) + H^(15/4) * d^(7/6) * K^(1/2) / ε).

differential privacy reinforcement learning linear MDPs regret minimization personalized decision-making

2025-04-12abs ↗pdf ↗

Actor Critic with Differentially Private Criticcs.LG

Reinforcement learning algorithms are known to be sample inefficient, and often performance on one task can be substantially improved by leveraging information (e.g., via pre-training) on other related tasks. In this work, we propose a technique to achieve such knowledge transfer in cases where agent trajectories conta…

2019-10-14abs ↗pdf ↗

Offline Reinforcement Learning for Human-Guided Human-Machine Interaction with Private Informationstat.ML

Study human-machine interaction with private info using offline RL.

problem Confounding bias and distributional mismatch in offline RL for human-guided interaction.

method Developed a novel identification result and OPE method to address confounding bias, and used pessimism to tackle distributional mismatch.

result Policy pair converges to optimal one at satisfactory rate under mild assumptions.

Human-machine interaction Offline reinforcement learning Confounding bias Distributional mismatch Instrumental variable

2022-12-23abs ↗pdf ↗

Strategic Decision-Making in the Presence of Information Asymmetry: Provably Efficient RL with Algorithmic Instrumentsstat.ML

New algorithm learns optimal policies in strategic MDPs with private types.

problem Optimal policy learning in strategic MDPs with private types and information asymmetry.

method PLAN algorithm using instrumental variable regression and pessimism principle.

result PLAN achieves near-optimal policy with

1 / \sqrt{K}

optimality.

reinforcement learning strategic decision-making information asymmetry instrumental variables myopic agents

2022-08-23abs ↗pdf ↗

Locally Differentially Private Reinforcement Learning for Linear Mixture Markov Decision Processescs.LG

Paper proposes a privacy-preserving RL algorithm for linear MDPs with theoretical guarantees.

problem Protecting users' private data in personalized services using RL.

method Local differential privacy (LDP) for RL with linear function approximation.

result Achieves a regret bound of $O(d^{5/4}H^{7/4}T^{3/4}\left(\log(1/δ) ight)^{1/4}\sqrt{1/\varepsilon})$ for linear mixture MDPs.

Reinforcement Learning Privacy Preserving Local Differential Privacy Linear Mixture MDPs Regret Analysis

2021-10-19abs ↗pdf ↗

Offline Reinforcement Learning with Differential Privacycs.LG

Offline RL algorithms protect privacy while learning from sensitive data.

problem Learning sensitive data in financial, legal, and healthcare applications.

method Differential privacy guarantees for offline RL algorithms.

result Provable prevention of privacy risks with strong learning bounds.

differential privacy offline reinforcement learning privacy protection sensitive data financial applications

2022-06-02abs ↗pdf ↗

Privacy Preserving Off-Policy Evaluationcs.LG

Many reinforcement learning applications involve the use of data that is sensitive, such as medical records of patients or financial information. However, most current reinforcement learning methods can leak information contained within the (possibly sensitive) data on which they are trained. To address this problem, w…

2019-02-01abs ↗pdf ↗

Differentially Private Meta-Learningcs.LG

Parameter-transfer is a well-known and versatile approach for meta-learning, with applications including few-shot learning, federated learning, and reinforcement learning. However, parameter-transfer algorithms often require sharing models that have been trained on the samples from specific tasks, thus leaving the task…

2019-09-12abs ↗pdf ↗

Multi-agent Deep Reinforcement Learning with Extremely Noisy Observationscs.LG

Multi-agent reinforcement learning systems aim to provide interacting agents with the ability to collaboratively learn and adapt to the behaviour of other agents. In many real-world applications, the agents can only acquire a partial view of the world. Here we consider a setting whereby most agents' observations are al…

2018-12-03abs ↗pdf ↗

Multi-Agent Reinforcement Learning via Double Averaging Primal-Dual Optimizationcs.LG

Despite the success of single-agent reinforcement learning, multi-agent reinforcement learning (MARL) remains challenging due to complex interactions between agents. Motivated by decentralized applications such as sensor networks, swarm robotics, and power grids, we study policy evaluation in MARL, where agents with jo…

2018-06-03abs ↗pdf ↗

Privately Learning Decision Lists and a Differentially Private Winnowcs.LG

New algorithms for privately learning decision lists and halfspaces.

problem Private learning of decision lists and halfspaces.

method Differentially private algorithms for PAC and online models.

result Private algorithms match or surpass non-private guarantees.

differential privacy private learning decision lists halfspaces Winnow algorithm

2026-02-07abs ↗pdf ↗

The Trade-Offs of Private Predictioncs.LG

This paper studies trade-offs in private prediction methods.

problem Leakage of training data information in machine learning predictions.

method Private training and private prediction methods with trade-offs.

result Private training methods outperform private prediction methods in various settings.

private prediction machine learning privacy trade-offs training methods

2020-07-09abs ↗pdf ↗

Polynomial Time and Private Learning of Unbounded Gaussian Mixture Modelsstat.ML

Private learning of Gaussian Mixture Models without boundedness assumptions.

problem Private estimation of parameters of Gaussian Mixture Models with unbounded components.

method Reduction to non-private problem, blackbox privatization, Moitra and Valiant's algorithm.

result First sample complexity upper bound and polynomial time algorithm for privately learning GMMs.

Gaussian Mixture Models Private Learning Differential Privacy Polynomial Time Unbounded Components

2023-03-07abs ↗pdf ↗

Limits of Private Learning with Access to Public Datacs.LG

We consider learning problems where the training set consists of two types of examples: private and public. The goal is to design a learning algorithm that satisfies differential privacy only with respect to the private examples. This setting interpolates between private learning (where all examples are private) and cl…

2019-10-25abs ↗pdf ↗

Differentially Private Learning Needs Better Features (or Much More Data)cs.LG

Private learning needs more data or better features.

problem Improving differentially private machine learning performance.

method Demonstrates the need for either more private data or better features.

result Private learning requires either more data or better features.

differential privacy machine learning feature learning data privacy

2020-11-23abs ↗pdf ↗

DADI: Dynamic Discovery of Fair Information with Adversarial Reinforcement Learningcs.LG

We introduce a framework for dynamic adversarial discovery of information (DADI), motivated by a scenario where information (a feature set) is used by third parties with unknown objectives. We train a reinforcement learning agent to sequentially acquire a subset of the information while balancing accuracy and fairness …

2019-10-30abs ↗pdf ↗

Closure Properties for Private Classification and Online Predictioncs.LG

Bounds on Littlestone dimension for private learning and online prediction.

problem Understanding the Littlestone dimension of composed classes for private learning.

method Deriving bounds on Littlestone dimension and transforming private learners.

result Improved bounds on sample complexity for private learning.

Littlestone dimension private learning online prediction aggregation rule composed class

2020-03-10abs ↗pdf ↗

Private Distribution Learning with Public Data: The View from Sample Compressioncs.LG

Private distribution learning with public data, leveraging sample compression schemes.

problem Private distribution learning with public and private samples under differential privacy constraints.

method Connection to sample compression schemes and list learning.

result At least d public samples are necessary for private learnability of Gaussians in R^d.

private distribution learning differential privacy sample compression list learning Gaussian distributions

2023-08-11abs ↗pdf ↗

Private Learning Implies Online Learning: An Efficient Reductioncs.LG

We study the relationship between the notions of differentially private learning and online learning in games. Several recent works have shown that differentially private learning implies online learning, but an open problem of Neel, Roth, and Wu \cite{NeelAaronRoth2018} asks whether this implication is {\it efficient}…

2019-05-27abs ↗pdf ↗

A Computational Separation between Private Learning and Online Learningcs.LG

New findings show limitations in converting private learning to online learning efficiently.

problem Limitations in converting private learning to online learning efficiently.

method Assuming one-way functions, we show an efficient conversion from pure-private learners to online learners is impossible.

result Efficient conversion from pure-private learners to online learners is impossible under certain assumptions.

private learning online learning one-way functions sample complexity computational efficiency

2020-07-11abs ↗pdf ↗

Oracle-Efficient Differentially Private Learning with Public Datastat.ML

Efficiently learns private models using public data.

problem Improving private learning performance with public data.

method Proves computationally efficient algorithms for private learning with public data.

result First computationally efficient algorithms for private learning with public data.

differential privacy private learning public data statistical tractability optimization oracle

2024-02-13abs ↗pdf ↗

Exploring the Limits of Differentially Private Deep Learning with Group-wise Clippingcs.LG

Improved differentially private deep learning with group-wise clipping techniques.

problem Efficiency and privacy trade-offs in deep learning models.

method Group-wise clipping techniques (per-layer and per-device) to reduce compute time and memory overhead.

result Private learning with group-wise clipping achieves similar or better performance than non-private learning with less wall time.

differentially private deep learning group-wise clipping privacy-utility trade-off GPT-3 per-layer clipping

2022-12-03abs ↗pdf ↗

Choosing Public Datasets for Private Machine Learning via Gradient Subspace Distancestat.ML

Algorithm selects public datasets for private machine learning.

problem Choosing the most suitable public dataset for private machine learning.

method Measures gradient subspace distance between public and private datasets.

result Excess risk scales with the subspace distance between gradients.

private machine learning gradient subspace distance differentially private stochastic gradient descent public datasets

2023-03-02abs ↗pdf ↗

Privacy-preserving Q-Learning with Functional Noise in Continuous State Spacesstat.ML

We consider differentially private algorithms for reinforcement learning in continuous spaces, such that neighboring reward functions are indistinguishable. This protects the reward information from being exploited by methods such as inverse reinforcement learning. Existing studies that guarantee differential privacy a…

2019-01-30abs ↗pdf ↗

When Does Differentially Private Learning Not Suffer in High Dimensions?cs.LG

Private learning can perform well in high dimensions, contrary to known results.

problem When does differentially private learning not suffer in high dimensions?

method Introduced a condition called restricted Lipschitz continuity to derive improved bounds for excess empirical and population risks.

result Gradients in private fine-tuning of large models are mostly controlled by a few principal components, similar to conditions for convex settings.

differential privacy convex learning high-dimensional models private fine-tuning gradient behavior

2022-07-01abs ↗pdf ↗

On the Benefits of Public Representations for Private Transfer Learning under Distribution Shiftcs.LG

Public pretraining improves private model training even in extreme distribution shift scenarios.

problem Improving private model training accuracy in settings with large distribution shift.

method Empirical evaluation and theoretical explanation of public representations improving private training accuracy.

result Public representations can improve private training accuracy by up to 67% over private training from scratch in settings with large distribution shift.

private transfer learning distribution shift public pretraining differential privacy sample complexity

2023-12-24abs ↗pdf ↗

Private Post-GAN Boostingcs.LG

Private PGB boosts synthetic data quality using GANs and privacy techniques.

problem Differentially private GANs struggle with convergence and poor output quality.

method Combines reweighted samples from GAN training using Private Multiplicative Weights method.

result Improves synthetic data quality across various datasets and tasks.

Differential Privacy GANs Post-GAN Boosting Private Multiplicative Weights Synthetic Data

2020-07-23abs ↗pdf ↗

Robust and Private Learning of Halfspacescs.LG

Study privacy and robustness in learning halfspaces, proving hard trade-offs.

problem Balancing privacy and robustness in learning halfspaces.

method Proves nearly tight bounds on sample complexity for robust private learning of halfspaces.

result Robust and private learning is harder than robust or private learning alone.

Privacy Robustness Halfspaces Differential Privacy Adversarial Robustness

2020-11-30abs ↗pdf ↗

Differentially-Private Federated Linear Banditscs.LG

Paper develops a private algorithm for multi-agent learning in bandits.

problem Private cooperative learning in decentralized systems.

method Developed extsc{FedUCB} algorithm for multi-agent learning.

result Improves pseudoregret bounds and empirical performance.

differentially-private federated learning bandits multi-agent cooperative learning

2020-10-22abs ↗pdf ↗

Private Online Learning via Lazy Algorithmscs.LG

New private algorithms for online learning improve regret in high privacy regimes.

problem Private online learning from experts and convex optimization.

method Transformed lazy algorithms for differential privacy.

result Improved regret bounds for DP-OPE and DP-OCO.

private online learning differential privacy online prediction online convex optimization lazy algorithms

2024-06-05abs ↗pdf ↗

Private Learning with Public Featurescs.LG

New private learning algorithms improve utility in tasks with public features.

problem Private learning with public features in recommendation and ad prediction.

method Developed algorithms that protect only certain sufficient statistics, improving utility for linear regression and private recommendation benchmarks.

result Achieved state-of-the-art performance on private recommendation benchmarks.

private learning public features recommendation ad prediction sufficient statistics

2023-10-24abs ↗pdf ↗

Differentially Private Hyperparameter Tuning using Local Bayesian Optimizationstat.ML

Differentially private hyperparameter tuning improves privacy in machine learning.

problem Hyperparameter tuning leaks private information through selected configurations.

method Local Bayesian optimization using Gaussian Process surrogate for private gradient approximation.

result DP-GIBO converges to locally optimal hyperparameters with polynomial dimensional dependence.

differential privacy hyperparameter tuning local Bayesian optimization Gaussian Process machine learning

2025-02-09abs ↗pdf ↗

A Dynamic Model of Private Asset Allocationq-fin.PM

Dynamic model considers private asset markets' complexities.

problem Understanding and optimizing private asset allocation.

method State-of-the-art dynamic model with machine learning.

result Optimal investment policies quantified over fund life.

private assets asset allocation machine learning regulatory constraints capital charges

2025-03-03abs ↗pdf ↗

Differentially Private Image Classification by Learning Priors from Random Processescs.CV

DP-RandP improves privacy-utility tradeoff in DP-SGD by learning priors from random processes.

problem Improving the performance of differentially private stochastic gradient descent (DP-SGD) on private data.

method A three-phase approach that learns priors from images generated by random processes and transfers these priors to private data.

result New state-of-the-art accuracy on CIFAR10, CIFAR100, MedMNIST, and ImageNet for various privacy budgets.

differentially private image classification stochastic gradient descent random processes privacy-utility tradeoff

2023-06-08abs ↗pdf ↗

DePAint: A Decentralized Safe Multi-Agent Reinforcement Learning Algorithm considering Peak and Average Constraintscs.MA

DePAint solves MARL for agents with local constraints, privacy, and no central controller.

problem Training multi-agent systems to optimize rewards while adhering to safety constraints in a decentralized setting.

method Formulated as a decentralized constrained multi-agent Markov Decision Problem, proposed DePAint method using momentum-based decentralized policy gradient.

result First privacy-preserving fully decentralized MARL algorithm considering both peak and average constraints.

multi-agent reinforcement learning decentralized privacy-preserving policy gradient reinforcement learning constraints

2023-10-22abs ↗pdf ↗

PILLAR: How to make semi-private learning more effectivecs.LG

PILLAR improves SP learning with less private data.

problem Efficiently learning with semi-private data under privacy constraints.

method Uses pre-trained public data features to reduce private data requirements.

result Significantly lower private labelled sample complexity achieved.

semi-supervised learning semi-private learning privacy constraints public data features sample complexity reduction

2023-06-06abs ↗pdf ↗

How unfair is private learning ?cs.LG

Private learning is hard when data is long-tailed.

problem Achieving both privacy and fairness in machine learning with long-tailed data.

method Theoretical analysis and experimental validation on various datasets and algorithms.

result Relaxing overall accuracy can lead to good fairness even with strict privacy requirements.

private learning fairness long-tailed data machine learning privacy

2022-06-08abs ↗pdf ↗

Learning from Mixtures of Private and Public Populationscs.LG

New privacy-preserving learning model for mixtures of private and public data.

problem Learning from datasets with both private and public data, where privacy concerns differ.

method Designing a differential privacy-preserving learning algorithm for a mixture of private and public sub-populations.

result Linear classifiers can be learned with sample complexity comparable to non-private PAC-learning, even when privacy status correlates with labels.

Privacy-preserving learning Differential privacy Mixture of private and public data Linear classifiers PAC-learning

2020-08-01abs ↗pdf ↗

Private PAC learning implies finite Littlestone dimensioncs.LG

We show that every approximately differentially private learning algorithm (possibly improper) for a class $H$ with Littlestone dimension~ $d$ requires $Ω\bigl(\log^*(d)\bigr)$ examples. As a corollary it follows that the class of thresholds over $\mathbb{N}$ can not be learned in a private manner; this resolves open qu…

2018-06-04abs ↗pdf ↗

Scalable and Provably Accurate Algorithms for Differentially Private Distributed Decision Tree Learningcs.LG

Paper introduces algorithms for private decision tree learning.

problem Private decision tree learning in distributed settings.

method Proposes DP-TopDown, NoisyCounts, and LocalRNM.

result First utility guarantees for private decision tree learning.

differential privacy decision trees distributed learning top-down privacy-preserving algorithms

2020-12-19abs ↗pdf ↗

Optimal differentially private kernel learning with random projectionstat.ML

Optimizes differentially private kernel learning with random projection.

problem Privacy-preserving learning algorithms with optimal performance.

method Differentially private kernel ERM algorithm based on random projection in reproducing kernel Hilbert space.

result Achieves minimax-optimal excess risk rates for various loss functions.

differential privacy kernel learning random projection empirical risk minimization minimax-optimal

2025-07-23abs ↗pdf ↗