Bryan Kian Hsiang Low

About

Bryan Kian Hsiang Low is a leading researcher in artificial intelligence, with key contributions spanning Bayesian optimization, inverse reinforcement learning (IRL), and risk-aware decision-making. His work addresses fundamental challenges in learning from demonstration and optimizing under uncertainty. In a highly cited 2020 paper, Low tackled the ill-posed nature of IRL by introducing a Bayesian optimization framework to efficiently explore the space of reward functions, enabling more robust value alignment and robot learning from demonstration. This work has garnered 7 citations and is foundational for subsequent IRL research. More recently, in 2021, Low developed the first no-regret algorithm for value-at-risk (VaR) optimization using Gaussian processes, a breakthrough for critical applications where risk assessment is paramount. His V-UCB algorithm provides provable guarantees for maximizing VaR of black-box functions, with 3 citations already. Low’s research is characterized by rigorous theoretical foundations and practical impact, bridging the gap between optimization theory and real-world deployment. His work continues to influence fields such as robotics, finance, and autonomous systems, making him a key figure in advancing safe and efficient AI.

Research Focus

Key Achievements

Top Papers

1. 1
   Efficient Exploration of Reward Functions in Inverse Reinforcement\n Learning via Bayesian Optimization

   7 citations · 2020
2. 2
   Value-at-Risk Optimization with Gaussian Processes

   3 citations · 2021

Key Collaborators

Contact & Links