We propose Fine-Tuning with Lossy Affordance Planner (FLAP), a framework that
leverages diverse offline data for learning representations, goal-conditioned policies, and affordance
models that enable rapid fine-tuning to new tasks in target scenes.
We propose a new form of state abstraction called goal-conditioned bisimulation that captures functional equivariance, allowing for the reuse of skills to achieve new goals in goal-conditioned reinforcement learning.
Machine Learning / Computer Vision Intern, Ambi Robotics
January - May 2022
Worked on improving the computer vision system which powers AmbiSort.
Software Engineering Intern, UiPath
June - August 2021
Worked on the Insights Team, which tracks UiPath's entire robotic process automation program.
Notes: Some of my notes on machine learning, math, etc.
Coursework: My coursework as an undergraduate at UC Berkeley.
Projects: Coding projects I have done in the past.