Research Projects

My research interests and projects (Updated August 2018) .


  1. Modeling DA neurons with trial-by-trial Q learning model (2018 - current)
    • First-year Spring Rotation Project (PIs: Nathaniel Daw and Ilana Witten, Mentors: Marcelo Mattar and Nathan Parker)
    • Previous work from Witten lab has shown that DA neurons with dorsal projections encode movement while ventral projections encode reward prediction error (rpe) in rodents [ref]
    • We built a Q learning model using the data from the rodent's behavior, then used the Q values to determine if the dorsal terminal signals also encode some form of rpe. 
    • Manuscript in Preperation + Poster in preperation for SfN 2018. 
  2. Deep Learning Framework for Perceptual Learning (2012 - 2015)
    • Undergraduate Thesis (PI: Prof. Jay McClelland, Mentor: Dr. Andrew Saxe)
    • Presented a deep learning framework for perceptual learning, arguing that the neural mechanisms of orientation discrimination can be modeled with a deep, chain-like network initialized to mimic the visual system and trained with back-propagation gradient descent.
    • Developed a deep neural network in MATLAB to model results from several neurobiology papers.
    • Evaluated analytical results from a simplified linear deep neural network to derive novel predictions. 
    • Results presented at COGSCI 2014 and COSYNE 2015
    • Manuscript in Preperation 



  1. Analyzing Sketch Features with Deep Neural Networks (Princeton, 2018)
    • First-year Fall Semester Rotation Project in Prof. Ken Norman's lab (PI/Project Lead: Dr. Judy Fan)
  2. Modeling Spontaneous Hypothesis Generation with Particle Filter (MIT, 2015)
    • Summer Internship Project as a Visting Student in Prof. Josh Tenenbaum's lab (PI/Advisor: Prof. Sam Gershman) 
    • Examined a computational model that describes hypothesis generation processes such as doctors choosing the appropriate diagnostic hypothesis when presented with a series of input observation
    • Developed probabilistic model of particle filter to capture the quantitative computation involved in generating hypotheses spontaneously 


I am a Computational Cognitive Neuroscientist who focuses on modeling and has strong interests in Theoretical Neuroscience. I was trained in Cognitive Modeling (specifically, connectionist modeling), and I am currently pursuing a PhD degree in Neuroscience at Princeton as part of the Daw Lab

I enjoy working on interdisciplinary research problems that study cognitive behaviors by modeling neuroscience data. My research interests are broad because I don't have a particular preference for the cognitive behavior I study, the type of dataset I model, or the modeling methodology. I am excited by modeling work that have interesting theoretical implications for neuroscience and/or cognitive science. These days I prefer working with rodent behavioral + neural data. I have a strong interest in doing more theoretical work.

My most extensive project was my undergraduate thesis that involved deep neural network theory and modeling perceptual learning. I consider myself fairly proficient at building deep neural networks, more so in theory than in applications. At Princeton, I plan on focusing more on reinforcement learning models. I also have general interests in AI/ML applications of brain inspired models. As a former product manager who worked in tech, I am always excited by work that can impact or apply to fields outside of my immediate field. 


A smattering of random projects outside of academia that I've done.

1. After college, I spent about two years pursuing Product/Program Management in the Bay Area. I interned at Tableau Softwares (2015 Summer). I worked for Microsoft Powerpoint for a full year (2016 - 2017). My main project for MSFT PPT was Accessibility i.e. I helped develop features that help blind and deaf users use MSFT PPT effectively. Microsoft subsequently won the Helen Keller Achievement Award from the American Foundation for the Blind in 2018 for their contributions to accessible technology. 

2. In highschool, I spent several summers working for Stanford's REAP Program to help rural children in China have better access to education.