Federated linear contextual bandits with user-level differential privacy

R Huang, H Zhang, L Melis, M Shen… - International …, 2023 - proceedings.mlr.press
This paper studies federated linear contextual bandits under the notion of user-level
differential privacy (DP). We first introduce a unified federated bandits framework that can …

High dimensional differentially private stochastic optimization with heavy-tailed data

L Hu, S Ni, H Xiao, D Wang - Proceedings of the 41st ACM SIGMOD …, 2022 - dl.acm.org
As one of the most fundamental problems in machine learning, statistics and differential
privacy, Differentially Private Stochastic Convex Optimization (DP-SCO) has been …

Collaborative ai teaming in unknown environments via active goal deduction

Z Zhang, H Zhou, M Imani, T Lee, T Lan - arXiv preprint arXiv:2403.15341, 2024 - arxiv.org
With the advancements of artificial intelligence (AI), we're seeing more scenarios that require
AI to work closely with other agents, whose goals and strategies might not be known …

Stochastic graphical bandits with heavy-tailed rewards

Y Gou, J Yi, L Zhang - Uncertainty in Artificial Intelligence, 2023 - proceedings.mlr.press
We consider stochastic graphical bandits, where after pulling an arm, the decision maker
observes rewards of not only the chosen arm but also its neighbors in a feedback graph …

Distributed differential privacy in multi-armed bandits

SR Chowdhury, X Zhou - arXiv preprint arXiv:2206.05772, 2022 - arxiv.org
We consider the standard $ K $-armed bandit problem under a distributed trust model of
differential privacy (DP), which enables to guarantee privacy without a trustworthy server …

Byzantine-robust federated linear bandits

A Jadbabaie, H Li, J Qian, Y Tian - 2022 IEEE 61st Conference …, 2022 - ieeexplore.ieee.org
In this paper, we study a linear bandit optimization problem in a federated setting where a
large collection of distributed agents collaboratively learn a common linear bandit model …

Private stochastic optimization with large worst-case lipschitz parameter: Optimal rates for (non-smooth) convex losses and extension to non-convex losses

A Lowy, M Razaviyayn - International Conference on …, 2023 - proceedings.mlr.press
We study differentially private (DP) stochastic optimization (SO) with loss functions whose
worst-case Lipschitz parameter over all data points may be extremely large. To date, the vast …

On the statistical complexity of estimation and testing under privacy constraints

C Lalanne, A Garivier, R Gribonval - arXiv preprint arXiv:2210.02215, 2022 - arxiv.org
The challenge of producing accurate statistics while respecting the privacy of the individuals
in a sample is an important area of research. We study minimax lower bounds for classes of …

On private and robust bandits

Y Wu, X Zhou, Y Tao, D Wang - Advances in Neural …, 2024 - proceedings.neurips.cc
We study private and robust multi-armed bandits (MABs), where the agent receives Huber's
contaminated heavy-tailed rewards and meanwhile needs to ensure differential privacy. We …

Decentralized randomly distributed multi-agent multi-armed bandit with heterogeneous rewards

M Xu, D Klabjan - Advances in Neural Information …, 2024 - proceedings.neurips.cc
We study a decentralized multi-agent multi-armed bandit problem in which multiple clients
are connected by time dependent random graphs provided by an environment. The reward …