1 citations · 3 across the 20 of their papers we have counts for
14 papers · 1 filter
ProbGuard: Calibrated Safety Risk Estimation from LLM Output Distributions
Xinzhe Huang, Biwu Yao, Kedong Xiu +4
Recent research on Large Language Model (LLM) safety has widely adopted guardrails to identify unsafe LLM outputs. Existing guardrails typically formulate safety assessment as a de…
H+: An Efficient Similarity-Aware Aggregation for Byzantine Resilient Federated Learning
Shiyuan Zuo, Rongfei Fan, Cheng Zhan +3
Federated Learning (FL) enables decentralized model training without sharing raw data. However, it remains vulnerable to Byzantine attacks, which can compromise the aggregation of…
Consistent Estimation of Numerical Distributions under Local Differential Privacy by Wavelet Expansion
Puning Zhao, Zhikun Zhang, Bo Sun +4
Distribution estimation under local differential privacy (LDP) is a fundamental and challenging task. Significant progresses have been made on categorical data. However, due to dif…
On Theoretical Limits of Learning with Label Differential Privacy
Puning Zhao, Chuan Ma, Li Shen +2
Label differential privacy (DP) is designed for learning problems involving private labels and public features. While various methods have been proposed for learning under label DP…
Differential Private Stochastic Optimization with Heavy-tailed Data: Towards Optimal Rates
Puning Zhao, Jiafei Wu, Zhe Liu +3
We study convex optimization problems under differential privacy (DP). With heavy-tailed gradients, existing works achieve suboptimal rates. The main obstacle is that existing grad…
Sequential Federated Learning in Hierarchical Architecture on Non-IID Datasets
Xingrun Yan, Shiyuan Zuo, Rongfei Fan +4
In a real federated learning (FL) system, communication overhead for passing model parameters between the clients and the parameter server (PS) is often a bottleneck. Hierarchical…
Contextual Bandits for Unbounded Context Distributions
Puning Zhao, Rongfei Fan, Shaowei Wang +4
Nonparametric contextual bandit is an important model of sequential decision making problems. Under -Tsybakov margin condition, existing research has established a regret bound…
Efficient Federated Learning against Byzantine Attacks and Data Heterogeneity via Aggregating Normalized Gradients
Shiyuan Zuo, Xingrun Yan, Rongfei Fan +4
Federated Learning (FL) enables multiple clients to collaboratively train models without sharing raw data, but is vulnerable to Byzantine attacks and data heterogeneity, which can…
Enhancing Learning with Label Differential Privacy by Vector Approximation
Puning Zhao, Rongfei Fan, Huiwen Wu +3
Label differential privacy (DP) is a framework that protects the privacy of labels in training datasets, while the feature vectors are public. Existing approaches protect the priva…
CG-FedLLM: How to Compress Gradients in Federated Fune-tuning for Large Language Models
Huiwen Wu, Xiaogang Xu, Deyi Zhang +3
The success of current Large-Language Models (LLMs) hinges on extensive training data that is collected and stored centrally, called Centralized Learning (CL). However, such a coll…
Emulating Full Participation: An Effective and Fair Client Selection Strategy for Federated Learning
Qingming Li, Juzheng Miao, Puning Zhao +5
In federated learning, client selection is a critical problem that significantly impacts both model performance and fairness. Prior studies typically treat these two objectives sep…
A Huber Loss Minimization Approach to Mean Estimation under User-level Differential Privacy
Puning Zhao, Lifeng Lai, Li Shen +3
Privacy protection of users' entire contribution of samples is important in distributed systems. The most effective approach is the two-stage scheme, which finds a small interval f…
Soft Label PU Learning
Yong Lei, Puning Zhao, Jintao Deng +1
PU learning refers to the classification problem in which only part of positive samples are labeled. Existing PU learning methods treat unlabeled samples equally. However, in many…
Federated Learning Resilient to Byzantine Attacks and Data Heterogeneity
Shiyuan Zuo, Xingrun Yan, Rongfei Fan +4
This paper addresses federated learning (FL) in the context of malicious Byzantine attacks and data heterogeneity. We introduce a novel Robust Average Gradient Algorithm (RAGA), wh…