Student Innovator Spotlight: Vincent Liu

A recipient of the Borealis AI 2021 Fellowship discusses his work in reinforcement learning.


Tell us about your research!

I am interested in sequential decision-making for real world problems. More specifically, my research interests lie in reinforcement learning (RL) and bandits. In many real world applications, we often have offline and observational data that we can leverage to improve RL and bandits. 

The basic concept of RL and bandits is that we aim to learn a policy that chooses an action or a sequence of actions to maximize a notion of rewards. AlphaGo, a computer program that plays the game of Go by maximizing the chance of winning, is a well-known example of a real world RL application. 

How has the Borealis AI graduate fellowship supported your work?

The Borealis AI Fellowship has provided me with valuable encouragement and motivation to continue my research direction. This award also allowed me to focus on my current research project, which was eventually turned into a submission to a top AI conference. 

What does innovation mean to you?

To me, innovation (in scientific research) means introducing novel ideas, designing novel algorithms, or providing novel analysis for open and unsolved problems. I think an innovator not only has sufficient knowledge in a particular field, but also has the ability to think out of the box. 

What’s one big problem you want to solve through your work?

A big problem I want to solve is understanding when and whether offline RL algorithms work in realistic settings— for example, when the environment is partially observable or non-stationary— and develop better algorithms for these settings with provable guarantees. If we understand when our algorithms work, it is more practical to apply the algorithms in the real world.

What makes the U of A a great place to do your work and research?

U of A has world-class professors and researchers working on reinforcement learning and bandits. The professor and students at U of A also have good connections to the RL community around the world. I think U of A is a perfect place to do RL research or start a career as a RL researcher. 

Innovator Spotlight is a series that introduces you to students whose big ideas are making a big difference.

Do you know someone who’s breaking boundaries at the U of A? (Maybe it’s you!) We’re interested in hearing from students who are creating new solutions to make our world better, no matter their discipline or year of study.

Get in touch at

This interview has been edited for clarity.


About Vincent

Vincent is a third-year PhD student in the Department of Computing Science at the University of Alberta. He has a MSc. from the University of Alberta and B.B.A. from National Taiwan University.