Yan Sun
 |
Assistant Professor, |
About me
I am an assistant professor in the Department of Mathematical Science at New Jersey Institute of Technology. Prior to joining NJIT, I was a Postdoctoral Researcher in the Department of Statistics and Data Science at the University of Pennsylvania, jointly mentored by Prof. Edgar Dobriban, Prof. Ian Barnett, and Prof. Pratik Chaudhari, working on model calibration and uncertainty quantification of machine learning models. Before that, I was an applied scientist at Amazon.
I obtained my Ph.D. in 2022 from the Department of Statistics at Purdue University, supervised by Prof. Faming Liang. I received my bachelor’s degree in 2017 from the Department of Mathematics in Chu Kochen Honors College, Zhejiang University.
Research Interests
I am interested in connecting statistical theories with advanced Machine Learning models, and with theoretical guidance, develop effective methods to improve practice. Some topics that I have worked on include:Sparse Deep Learning: Consistency, High Dimensional Variable Selection, Network Pruning.
Bayesian Statistics: Bayesian Neural Network, Posterior Concentration, Markov Chain Monte Carlo.
Uncertainty Quantification: Model Calibration, Asymptotic Distribution.
Large Language Models: Watermarking, Sampling
Preprints
Foundations of Top-k Decoding For Language Models
Georgy Noarov, Soham Mallick, Tao Wang, Sunay Joshi, Yan Sun, Yangxinyu Xie, Mengxin Yu, Edgar Dobriban [PDF]An Analytical Characterization of Sloppiness in Neural Networks: Insights from Linear Models
Jialin Mao, Itay Griniasty, Yan Sun, Mark K Transtrum, James P Sethna, Pratik Chaudhari [PDF]Statistical Inference for Generative Model Comparison
Zijun Gao, Yan Sun [PDF], [Code]Watermarking language models with error correcting codes
Patrick Chao, Yan Sun, Edgar Dobriban, Hamed Hassani [PDF]A Confidence Interval for the \( {\ell_2} \) Expected Calibration Error
Yan Sun, Pratik Chaudhari, Ian Barnett, Edgar Dobriban [PDF], [Code]
Publications
Magnitude Pruning of Large Pretrained Transformer Models with a Mixture Gaussian Prior
Mingxuan Zhang, Yan Sun, Faming Liang Journal of Data Science (2024). [PDF]Extended fiducial inference: toward an automated process of statistical inference
Faming Liang, Sehwan Kim, Yan Sun Journal of the Royal Statistical Society: Series B (Statistical Methodology) (2024). [PDF]Deep network embedding with dimension selection
Tianning Dong, Yan Sun, Faming Liang Neural Networks (2024). [PDF]Sparse Deep Learning for Time Series Data: Theory and Applications
Mingxuan Zhang, Yan Sun, Faming Liang Advances in neural information processing systems (2023). [PDF]Nonlinear Sufficient Dimension Reduction with a Stochastic Neural Network
Siqi Liang, Yan Sun, Faming Liang Advances in neural information processing systems (2022). [PDF], [Code]A Kernel-Expanded Stochastic Neural Network.
Yan Sun, Faming Liang Journal of the Royal Statistical Society: Series B (Statistical Methodology) (2022). [PDF], [Code]Sparse Deep Learning: A New Framework Immune to Local Traps and Miscalibration.
Yan Sun, Wenjun Xiong, Faming Liang Advances in neural information processing systems (2021). [PDF], [Code]Learning Sparse Deep Neural Networks with Spike-and-Slab Priors.
Yan Sun, Qifan Song, Faming Liang Statistics & Probability Letters (2021). [PDF], [Code]Consistent Sparse Deep Learning: Theory and Computation.
Yan Sun*, Qifan Song*, Faming Liang Journal of the American Statistical Association (2021). [PDF], [Code]Extended stochastic gradient Markov chain Monte Carlo for large-scale Bayesian variable selection.
Qifan Song, Yan Sun, Mao Ye, Faming Liang Biometrika, (2020) [PDF]Variable selection via penalized neural network: a drop-out-one loss approach.
Mao Ye*, Yan Sun* International Conference on Machine Learning (2018). [PDF]