Dr. Qi (Roger) She
I am currently a research scientist in Intel Labs, China focusing on statistical machine learning, deep learning with applications in robotic vision. I obtained my Ph.D. degree from the Department of Electronic Engineering at the City University of Hong Kong where I was advised by Prof. Rosa H.M. Chan. During 2017, I was a Visiting Student Research Collaborator (VSRC) at Princeton University, advised by Prof. Jonathan Pillow. I also worked as a research intern in Tencent AI Lab focusing on computer vision project.
Previously, my research focused on statistical machine learning methods to extract hidden structure from high-dimensional neural data, infer brain connectivity using fully & empirical Bayes, and complex network study on multiple brain regions. I am interested in studying how information is encoded, decoded, and processed in our brain. Currently, motivated from brain computing, I would prefer to develop lifelong/continual adapation agent that can shape a cultivated understanding of the world from the current scene and their previous knowledge via an autonomous lifelong development. The ongoing project is listed in "Lifelong Robotic Vision" project page.
Before directly pursuing my Ph.D., I completed my B.Eng. within 1% (2/230) in Information Engineering in September 2014, and my bachelor final year project ranked #1 out of 230.
Senior Research Scientist, 2018/8-Current
Tencent AI Lab
Research Intern, 2017/12-2018/3
Princeton Neuroscience Institute, Princeton University
Visiting Student, 2017/3-2017/12
Department of Electronic Engineering, City University of Hong Kong
Ph.D. Student, 2014/9-2018/8
Department of Information Engineering, Peking University
Exchange Student, 2013-2014
Nanjing University of Posts and Telecommunications
Extracting Interpretable Latent Structure from Neural Spike Trains
We show how to capture the intrinsic dynamics and learn a concise, structured, and interpretable latent space.
NIPS 2018 (submitted), ICASSP 2018, AAAI 2018
Network Modeling of Spike Counts
We show how to model over-dispersion of count data under population settings with Generalized Linear Models
Submitted to IEEE Transactions on Signal Processing
Graph-theoretic Analysis of Multi-neuronal recordings
We apply graph theory to analyze the latent structure of neural data., giving insight into the graph-structured time series data.
Neural Dyanamics Modeling Using Kernal Generalized Linear Models
We show how to capture linear-nonlinear neuronal interactions with kernel-GLMs
EMBC 2015, 2016, 2018
Time-varying network modeling using Polya-Gamma augmentation
Oral presentation for the 10th Annual Student Competition of IEEE-EMBS HK-Macau Joint Chapter
“Reduced-Rank Linear Dynamical Systems”, Spotlight talk of AAAI2018 at Hilton New Orleans River-
side, New Orleans, Louisiana, USA, Feb. 2018
“Explicit and Implicit Networks of Neural Ensembles", the Poster presentation of Research Symposium for MSc and Research Students at City U., Nov. 2016
Photos of My Life
起初因为想锻炼身体爱上了这种运动，后来渐渐发现这种运动带给自己的不仅仅是健硕的身体，更多的是精疲力竭后内心的平静，什么都不想的感觉。Swimming is really wonderful !