Hi! I’m a Ph.D. candidate at Worcester Polytechnic Institute advised by Elke Rundensteiner and Xiangnan Kong in the Data Science program, a part of the Computer Science department. I am looking to combine reinforcement learning with sequence modeling to train computers to perceive the world like humans do and solve problems in healthcare and sustainability.
My CV can be found here.
- June 04, 2020: Talk at FSU on Early Classification.
- May 15, 2020: New KDD paper on Early Multi-label Classification.
- April 04, 2020: New ACL paper on the interpretability of attention mechanisms for text classification.
- Feb 26, 2020: New HEALTHINF paper on interpretable machine learning for C. Diff. detection. Won best poster.
- Nov 19, 2019: Talk at WPI Colloquium on Conditional Computation in RNNs.
- Oct 16, 2019: Two new papers at IEEE BigData on hierarchical RNNs for classifying series of clinical notes and trainable decay in attention mechanisms for clinical notes.
- Sep 16, 2019: Talk at University of Minnesota on Early Classification.
- May 10, 2019: Talk at Northeastern University's New England Machine Learning day poster session.
- April 29, 2019: New KDD paper on Early Classification.
- Time series modelling
- Recurrent Neural Networks
- Early Classification
- Healthcare applications
I am generally interested in sequence modeling, or building vector representations that capture relevant temporal dynamics in sequential data such as time series or text. So far, I have studied and published on early classification, clinical note classification, attention mechanisms for RNNs, meta word embeddings, interpretable machine learning, and irregularly-sampled time series. Much of my work involves Reinforcement Learning. On my Publications page I describe some of my representative research in more detail. I am also interested in applications to problems of human health and sustainability (particularly conservation biology).
Please feel free to contact me with questions regarding my research or our program at WPI. I am always looking to chat about research!