no code implementations • 1 Apr 2024 • Xiang Li, Feng Ruan, Huiyuan Wang, Qi Long, Weijie J. Su
In particular, we derive optimal detection rules for these watermarks under our framework.
no code implementations • 2 May 2023 • Zhiqi Bu, Zongyu Dai, Yiliang Zhang, Qi Long
Multiple imputation (MI) has been widely applied to missing value problems in biomedical, social and econometric research, in order to avoid improper inference in the downstream data analysis.
1 code implementation • 23 Nov 2022 • Zongyu Dai, Zhiqi Bu, Qi Long
Single imputation methods such as matrix completion methods do not adequately account for imputation uncertainty and hence would yield improper statistical inference.
no code implementations • 1 Jul 2022 • Changgee Chang, Zhiqi Bu, Qi Long
We provide theoretical investigation for the asymptotic properties of the proposed method for statistical inference as well as differential privacy, and evaluate its performance in simulations and real data analyses in comparison with several recently developed methods.
no code implementations • 7 Mar 2022 • Kan Chen, Qishuo Yin, Qi Long
Motivated by the theoretical analysis, we propose a novel objective function for estimating the ATE that uses the energy distance balancing score and hence does not require correct specification of the propensity score model.
no code implementations • 21 Dec 2021 • Zongyu Dai, Zhiqi Bu, Qi Long
Missing data are present in most real world problems and need careful handling to preserve the prediction accuracy and statistical consistency in the downstream analysis.
no code implementations • NeurIPS 2021 • Yiliang Zhang, Qi Long
When the goal is to develop a fair algorithm in the complete data domain where there are no missing values, an algorithm that is fair in the complete case domain may show disproportionate bias towards some marginalized groups in the complete data domain.
no code implementations • 22 Oct 2021 • Yiliang Zhang, Qi Long
Missing data are ubiquitous in the era of big data and, if inadequately handled, are known to lead to biased findings and have deleterious impact on data-driven decision makings.
no code implementations • 18 Jul 2021 • Qiyiwen Zhang, Zhiqi Bu, Kan Chen, Qi Long
Interestingly, we show a new equivalence between DP-SGD and DP-SGLD, implying that some non-Bayesian DP training naturally allows for uncertainty quantification.
1 code implementation • 15 Jun 2021 • Zhiqi Bu, Hua Wang, Zongyu Dai, Qi Long
Differentially private (DP) training preserves the data privacy usually at the cost of slower convergence (and thus lower accuracy), as well as more severe mis-calibration than its non-private counterpart.
no code implementations • 2 Mar 2021 • Shuxiao Chen, Qinqing Zheng, Qi Long, Weijie J. Su
A widely recognized difficulty in federated learning arises from the statistical heterogeneity among clients: local datasets often come from different but not entirely unrelated distributions, and personalization is, therefore, necessary to achieve optimal results from each individual's perspective.
1 code implementation • 22 Feb 2021 • Qinqing Zheng, Shuxiao Chen, Qi Long, Weijie J. Su
Federated learning (FL) is a training paradigm where the clients collaboratively learn models by repeatedly sharing information without compromising much on the privacy of their local sensitive data.
1 code implementation • 29 Jan 2021 • Cong Fang, Hangfeng He, Qi Long, Weijie J. Su
More importantly, when moving to the imbalanced case, our analysis of the Layer-Peeled Model reveals a hitherto unknown phenomenon that we term \textit{Minority Collapse}, which fundamentally limits the performance of deep learning models on the minority classes.
no code implementations • 1 Jan 2021 • Yiliang Zhang, Qi Long
While there is a growing body of literature on fairness in analysis of fully observed data, there has been little work on investigating fairness in analysis of incomplete data when the goal is to develop a fair algorithm in the complete data domain where there are no missing values.
no code implementations • 7 Aug 2020 • Kefei Liu, Qi Long, Li Shen
The sparse canonical correlation analysis (SCCA) is a bi-multivariate association model that finds sparse linear combinations of two sets of variables that are maximally correlated with each other.
1 code implementation • ICML 2020 • Qinqing Zheng, Jinshuo Dong, Qi Long, Weijie J. Su
To address this question, we introduce a family of analytical and sharp privacy bounds under composition using the Edgeworth expansion in the framework of the recently proposed f-differential privacy.
3 code implementations • 26 Nov 2019 • Zhiqi Bu, Jinshuo Dong, Qi Long, Weijie J. Su
Leveraging the appealing properties of $f$-differential privacy in handling composition and subsampling, this paper derives analytically tractable expressions for the privacy guarantees of both stochastic gradient descent and Adam used in training deep neural networks, without the need of developing sophisticated techniques as [3] did.