Ph.D. in Computation and Neural Systems
Caltech · Pasadena, CA
Thesis: Computational modeling of behavioral patterns and neural mechanisms of rapid task switching
I am a PhD student in Computation and Neural Systems at
Caltech, co-advised by Markus Meister and Richard Murray. My research is on computational modeling of rapid
task switching. Putting it simply, we put electrodes in
human brains while they are playing games and see what
the neurons are doing. My research is supported by NSERC
Postgraduate Scholarship-Doctoral (PGS-D).
I completed my BASc in Engineering Science at the University
of Toronto, majoring in Robotics and Bioengineering. I had
the previlage of working with
Michael Hoffman, Juan Álvaro Gallego, and Taufik Valiante on projects spanning bioinformatics, brain-computer interfaces,
and spiking neural networks. Beyond research, I worked as
a software engineer at Amazon and a game dev at a startup.
I enjoy playing badminton and snowboarding in my spare time.
Feel free to contact me via email: yefan2718 at gmail dot
com
Caltech · Pasadena, CA
Thesis: Computational modeling of behavioral patterns and neural mechanisms of rapid task switching
University of Toronto · Toronto, ON
cGPA: 3.96/4.00
Caltech · Pasadena, CA
Recording from populations of human neurons in epilepsy patients while they perform a driving task, and develop control-theoretic and deep learning models to uncover the neural mechanisms of rapid task switching.
Caltech · Pasadena, CA
Used GLM analysis and convolutional autoencoders with CCA to investigate the relationship between LFP and BOLD signals, discovering that gamma-band oscillatory power correlates best with BOLD at r = 0.7.
Neuron to Brain Lab · University of Toronto · Toronto, Canada
Used spiking neural network simulations to investigate how network inputs affect neuronal intrinsic plasticity and how neuron heterogeneity contributes to epileptic dynamics.
Be.Neuro Lab · Imperial College London · London, UK
Built a real-time closed-loop BCI for mouse experiments (40ms latency), optimized spike-detection algorithms for a 70% speedup, and developed LSTM-based neural decoders to predict hand velocity from population recordings.
Hoffman Lab · University of Toronto · Toronto, Canada
Developed a computational pipeline for epigenetic data processing and a new Poisson-based statistical model for peak-calling with biochemical calibration.
Preload · San Francisco, US
We use EMG to decode hand position and force, and sell the data to robotics companies. We are in YC F26.
Neuralink · South San Francisco, US
Worked on the Blindsight project.
Amazon · Toronto, Canada
Built a full-stack item details page for Amazon Kids' parent dashboard (React-Redux-TypeScript + Java) that shipped to 80k+ customers.
City From Naught · Toronto, Canada
Developed a 3D cyberpunk detective game in Unity (C#) released on Steam to a Very Positive rating (86%).
Final project for Caltech's intro to general relativity course. I summarize results of two papers discussing Morris-Thorne wormhole and possibility of using them for time travel.
Jan 2023 - Apr 2023
Collaborated in a team of four to design, build, and test an autonomous drone that traverses a path with obstacle avoidance.
My first side project. Created an application that can auto-generate Baroque-style dance music.