Research

Check out my Google Scholar profile for a full list of publications and patents.
Check out my Github for software.

Understanding neural population activity via machine learning
(Advised by Professor Byron Yu, Carnegie Mellon University)

My research focused on developing novel dimensionality reduction techniques to explain explain the high-dimensional structure and timecourse of neural population activity. In particular, we developed a novel probabilistic technique - Time-delay Gaussian-process Factor Analysis (TD-GPFA) that shows how modelling latent variables via Gaussian Processes allows us to tractably learn these delays over a continuous domain. We demonstrated our method on simulated data as well as real spike train recordings from neurons in a Macaque motor cortex.

See also:
Journal publication (Journal of Neural Computation (NECO), 2015)
Conference abstract (COSYNE, 2015)
Masters Thesis presentation (Robotics Institute, Carnegie Mellon University)
Software

Constraint Aware Robotic Cloth Folding
(Advised by Professor Pieter Abbeel, UC Berkeley)

We developed a technique to enable robots to fold a variety of clothing articles, using a cloth model that allows us to reason about the geometry rather than the physics of the cloth. Our work enables the robot to search for a time optimal solution with a richer set of primitives it can execute to complete user defined folds.

See also:
Conference Publication (ISER, 2012)
Videos

Other projects

Ballbot: A low-cost robot for tennis ball retrieval
(Advised by Professor Pieter Abbeel, UC Berkeley)

The aim was to build an intelligent, autonomous tennis ball-boy/girl created using low-cost, off-the-shelf equipment. I implemented search based motion planning algorithms for this project.

See also:
Technical Report

CINEMA CubeSat
(Space Sciences Lab, UC Berkeley)

This was a collaborative effort between Berkeley and South Korea to build and launch three Cubesats to provide cutting-edge magnetospheric science and critical space weather measurements. I helped develop initial Interface Control Documents and FPGA modules to interface the STEIN particle detector to the CubeSat.

See also:
Press release
Project website

Noodle (UC Berkeley)

As a winning entry to the 2011 University Mobile Challenge at the World Mobile Congress in Barcelona, our team pitched Noodle -- a cloud based Android application that enables students to easily customize, digitize and share notes.

See also:
Press release

Multicore Gridding for MRI image reconstruction
(Internship at General Electric Global Research, Bangalore, India)

I helped parallelize gridding algorithms to reconstruct images for non-cartesian MRI.