Dr. Xin Ding

Filtering out unrealistic images from trained generative adversarial networks (GANs) has attracted considerable attention recently. Two density ratio based subsampling methods---Discriminator Rejection Sampling (DRS) and Metropolis-Hastings GAN (MH-GAN)---were recently proposed, and their effectiveness in improving GANs was demonstrated on multiple datasets. However, DRS and MH-GAN are based on discriminator based density ratio estimation (DRE) methods, so they may not work well if the discriminator in the trained GAN is far from optimal. Moreover, they do not apply to some GANs (e.g., MMD-GAN). In this paper, we propose a novel Softplus (SP) loss for DRE. Based on it, we develop a sample-based DRE method in a feature space learned by a specially designed and pre-trained ResNet-34 (DRE-F-SP). We derive the rate of convergence of a density ratio model trained under the SP loss. Then, we propose three different density ratio based subsampling methods (DRE-F-SP+RS, DRE-F-SP+MH, and DRE-F-SP+SIR) for GANs based on DRE-F-SP. Our subsampling methods do not rely on the optimality of the discriminator and are suitable for all types of GANs. We empirically show our subsampling approach can substantially outperform DRS and MH-GAN on a synthetic dataset and the CIFAR-10 dataset, using multiple GANs.

21 October 2022 in Expert Systems with Applications

Journal Article

Distilling and transferring knowledge via cGAN-generated samples for image classification and regression

Zuheng Xu

1 January 2017 in Electronic Journal of Statistics

Journal Article

Sparse transition matrix estimation for high-dimensional and locally stationary vector autoregressive models

Xin Ding

Ziyi Qiu

Xiaohui Chen

We consider the estimation of the transition matrix in the high-dimensional time-varying vector autoregression (TV-VAR) models. Our model builds on a general class of locally stationary VAR processes that evolve smoothly in time. We propose a hybridized kernel smoothing and $\ell^{1}$-regularized method to directly estimate the sequence of time-varying transition matrices. Under the sparsity assumption on the transition matrix, we establish the rate of convergence of the proposed estimator and show that the convergence rate depends on the smoothness of the locally stationary VAR processes only through the smoothness of the transition matrix function. In addition, for our estimator followed by thresholding, we prove that the false positive rate (type I error) and false negative rate (type II error) in the pattern recovery can asymptotically vanish in the presence of weak signals without assuming the minimum nonzero signal strength condition. Favorable finite sample performances over the $\ell^{2}$-penalized least-squares estimator and the unstructured maximum likelihood estimator are shown on simulated data. We also provide two real examples on estimating the dependence structures on financial stock prices and economic exchange rates datasets.

28 June 2019 in Data Science and Engineering

Journal Article

Similarity queries, including range queries and k nearest neighbor (kNN) queries, in metric spaces have applications in many areas such as multimedia retrieval, computational biology and location-based services. With the growing volumes of data, a distributed method is required. In this paper, we propose an Asynchronous Metric Distributed System (AMDS), to support efficient metric similarity queries in the distributed environment. AMDS uniformly partitions the data with the pivot-mapping technique to ensure the load balancing, and employs publish/subscribe communication model to asynchronous process large scale of queries. The employment of asynchronous processing model also improves robustness and efficiency of AMDS. In addition, we develop efficient similarity search algorithms using AMDS. Extensive experiments using real and synthetic data demonstrate the performance of metric similarity queries using AMDS. Moreover, the AMDS scales sublinearly with the growing data size.

About Me

Dr. Ding Xin, Associate Professor, and Master’s Supervisor, is a recipient of a national-level young talent program. He received his B.S. degree in Statistics from Anhui University in 2014, his M.S. degree in Statistics from the University of Illinois at Urbana-Champaign in 2016, and his Ph.D. degree in Statistics from the University of British Columbia in 2021, under the supervision of Professor Z. Jane Wang (Fellow of the Canadian Academy of Engineering and IEEE Fellow) and Professor William J. Welch (ASA Fellow). He joined the School of Artificial Intelligence at Nanjing University of Information Science & Technology in 2023. His research interests include probabilistic generative models, knowledge distillation, and AI for Science (e.g., healthcare and meteorology). He has led one Youth Program of the National Natural Science Foundation of China, one General Program of Jiangsu Provincial Basic Research Fund, one supporting project for recipients of Jiangsu’s major national talent program, one Nanjing Science and Technology Innovation Project for Returned Scholars (Category C), and one start-up funding project from Nanjing University of Information Science & Technology. As first author, he has published more than ten papers in top international journals and conferences such as IEEE TPAMI, IEEE TSP, ESWA, Neurocomputing, ICLR, and AAAI.

Dr. Xin Ding

Dr. Xin Ding

Research Focus & Metrics

Academic Career Length

Publication Type Breakdown

Representative Publications

About Me

Social Links