# Research

In: Proceedings of the 9th Asian Conference on Machine Learning (ACML’17), Seoul, Korea, 2017.

Here’s a link to the poster.

August 2016 - May 2017 at Purdue University

• Statistical analysis (SA) is a complex process to deduce population properties from analysis of data. It usually takes a well-trained analyst to successfully perform SA, and it becomes extremely challenging to apply SA to big data applications.
• We propose to use deep neural networks to automate the Statistical Analysis(SA) process.
• We propose to construct convolutional neural networks (CNNs) to perform automatic model selection and parameter estimation, two most important SA tasks.
• Simulation study shows that both the neural model selector and neural parameter estimator demonstrate excellent performances. The idea and proposed framework can be further extended to automate the entire SA process and have the potential to revolutionize how SA is performed in big data analytics.

• ## Inferring Spatial Organization of Individual Topologically Associated Domains via Piecewise Helical Model

June 2015 - August 2016 at Purdue University, presented at JSM 2016, Chicago

• Understand the 3D spatial organizations of chromosomes and functional implications of such structure.
• Propose new, parsimonious, easy to interpret models for reconstructing 3D chromosomal structure from Hi-C data to fully characterize the 3D chromosomal structure and its structural variations.
• Apply the method to high resolution Hi-C dataset generated from mouse embryonic stem cells.
• https://github.com/RSquared1427/phm
• ## Nonlinear partial differential equation, Harmonic analysis, Banach function spaces

• Showed the local existence and uniqueness of the solution in the generalized Feichtinger algebra $M^s_{(1,1)}$ $(s\geq 3/2)$ with sufficiently small initial data in $M^{3/2}_{(1,1)}(R^2)$.
• Moreover, showed the ill-posedness of the solutions in the sense that the solution map is not C3 if the spatial regularity is below $M^{3/2}_{1,1}$.