Hi! I'm Albert.

I recently defended my PhD thesis in Caltech's Computing and Mathematical Sciences Department advised by Yisong Yue. I design efficient algorithms for autonomy and machine learning, like

• achieving high physical dexterity with real-time visual detection,
• learning robust predictors for time series data, and
• making robots grab stuff with math and large-scale visual simulation.

I was also a student researcher at the Robotics and AI Institute (formerly the Boston Dynamics AI Institute), where I worked on robotic dexterity and reinforcement learning and was a major contributor of judo, an open-source package for robot planning.

Before Caltech, I received my BS from Berkeley and MS from Stanford.

In past lives, I dabbled in product design and shipped some parts in the Apple Watch Series 4. I was also a policy debater in high school, where I earned a few distinctions, including winning the Cal Invitational + the Cal Round Robin and the Texas State Championship.

Reach me at lialberthao[at]gmail[dot]com.