I am a Ph.D. in Computer Science Department at Worcester Polytechnic Institute (WPI), advised by Prof. Neil Heffernan. Before joining WPI, I got a master degree in Computer Science at South China University of Technology .

My research interests are in the application of machine learning (especially deep learning), contextual bandits, and domain adaptation. I have applied deep learning models to solving various tasks (e.g. time-series prediction, text classification with memory networks, estimation of treatment effects using domain adaptation).

During summer 2016, I interned at Philips Research North America advised by Sadid Hasan. During summer 2017, I interned at Amazon as Applied Scientist mentored by Aaron Fraenkel.

Applied Scientist Intern

Amazon (June 2017 - August 2017) - San Diego, CA

• Worked in Transaction Risk Management Systems (TRMS) group.
• Applied reinforcement learning algorithms.

Research Intern

Philips Research North America (May 2016 - August 2016) - Cambridge, MA

• Implemented Key-Value Memory Networks and Gated Attention Readers to classify diagnoses from medical notes
• Participated in TREC Clinical Decision Support Track 2016

Research Assistant

ASSISTments Lab (May 2014 - May 2018) - WPI

• Implemented a student model (knowledge tracing) with Long Short-Term Memory (LSTM) to predict the probability of a student making errors on given problems and measure student’s knowledge level on given skills.
• Built a model for automated assignment grading tasks using memory networks and achieved state-of-the-art results on the Kaggle ASAP dataset.
• Estimated the individual treatment effect for students using a deep learning model from the dataset collected from randomized control trials running inside ASSISTments.
• Finished the Comprehensive Exam on topics including Bayesian optimization, Gaussian process, aggregation of crowdsourced labels, and estimation of the treatment effect.