Kyle Hsu

kylehkhsu at gmail dot com

I'm a visiting student researcher at Berkeley AI Research working with Sergey Levine and Chelsea Finn in the Robotic AI & Learning Lab.

I'm pursuing my Bachelor's in Engineering Science (Major in Robotics Engineering) at the University of Toronto. Previously, I worked on abstraction-based controller synthesis with Rupak Majumdar at the Max Planck Institute for Software Systems. I've also spent time under Joyce Poon at the University of Toronto and Ming C. Wu at UC Berkeley.

CV  /  Google Scholar  /  LinkedIn  /  GitHub

Research

I've had the fortune of participating in a diverse range of interesting research projects with great mentors and company.

Unsupervised Learning via Meta-Learning
Kyle Hsu, Sergey Levine, Chelsea Finn
arXiv preprint, 2018
project page / code

We propose CACTUs, an unsupervised learning algorithm that leverages meta-learning techniques to learn to learn tasks constructed from unlabeled data. CACTUs leads to significantly more effective downstream learning and enables few-shot learning without requiring labeled meta-learning datasets.

Lazy Abstraction-Based Control for Reachability
*Kyle Hsu, Rupak Majumdar, Kaushik Mallik, Anne-Kathrin Schmuck
arXiv preprint, 2018
project page / demo / code

A third step towards scalable abstraction-based controller synthesis. Extending lazy abstraction-based control to handle reachability specifications requires carefully determining and processing frontier states, which are prime candidates for making progress in the synthesis.

Lazy Abstraction-Based Control for Safety Specifications
*Kyle Hsu, Rupak Majumdar, Kaushik Mallik, Anne-Kathrin Schmuck
IEEE Conference on Decision and Control (CDC), 2018 (to appear)
project page / code

A second step towards scalable abstraction-based controller synthesis. By entwining abstraction and synthesis, we enable on-demand construction of the abstractions instead of having to do so upfront and for the entire system.

Multi-Layered Abstraction-Based Controller Synthesis for Continuous-Time Systems
*Kyle Hsu, Rupak Majumdar, Kaushik Mallik, Anne-Kathrin Schmuck
Hybrid Systems: Computation and Control (HSCC), 2018

A first step towards scalable abstraction-based controller synthesis. Instead of letting bottleneck regions of the system dictate the sole resolution at which we operate, we first construct multiple abstractions of the system at various spatial and temporal granularity, then do synthesis by adaptively switching between these layers.

Germanium Wrap-Around Photodetectors on Silicon Photonics
Ryan Going, Tae Joon Seok, Jodi Loo, Kyle Hsu, Ming C. Wu
Optics Express, 2015

We present a new photodetector architecture based on coupling the silicon waveguide and germanium diode by wrapping the latter around the former instead of stacking them together. As a high school student, I contributed by testing fabricated samples.


*Alphabetical authorship. This guy makes a nice website.