8 papers
Feature Bagging Provides Stability
Yuheng Ma, Qiang Sun
We study feature bagging through the lens of algorithmic stability. Feature bagging is an ensemble strategy that aggregates base learners trained on randomly subsampled feature sub…
PrAda-GAN: A Private Adaptive Generative Adversarial Network with Bayes Network Structure
Ke Jia, Yuheng Ma, Yang Li +1
We revisit the problem of generating synthetic data under differential privacy. To address the core limitations of marginal-based methods, we propose the Private Adaptive Generativ…
Locally Private Nonparametric Contextual Multi-armed Bandits
Yuheng Ma, Feiyu Jiang, Zifeng Zhao +2
Motivated by privacy concerns in sequential decision-making on sensitive data, we address the challenge of nonparametric contextual multi-armed bandits (MAB) under local differenti…
Better Locally Private Sparse Estimation Given Multiple Samples Per User
Yuheng Ma, Ke Jia, Hanfang Yang
Previous studies yielded discouraging results for item-level locally differentially private linear regression with -sparsity assumption, where the minimax rate for sample…
ALTER: Augmentation for Large-Table-Based Reasoning
Han Zhang, Yuheng Ma, Hanfang Yang
While extensive research has explored the use of large language models (LLMs) for table-based reasoning, most approaches struggle with scalability when applied to large tables. To…
Locally Private Estimation with Public Features
Yuheng Ma, Ke Jia, Hanfang Yang
We initiate the study of locally differentially private (LDP) learning with public features. We define semi-feature LDP, where some features are publicly available while the remain…
Bagged Regularized -Distances for Anomaly Detection
Yuchao Cai, Hanfang Yang, Yuheng Ma +1
We consider the paradigm of unsupervised anomaly detection, which involves the identification of anomalies within a dataset in the absence of labeled examples. Though distance-base…
Optimal Locally Private Nonparametric Classification with Public Data
Yuheng Ma, Hanfang Yang
In this work, we investigate the problem of public data assisted non-interactive Local Differentially Private (LDP) learning with a focus on non-parametric classification. Under th…