Eugene Teoh


Hello! I was previously a Senior Machine Learning Research Engineer at a Robot Learning Lab in London, working with Stephen James. I work on research in Robot Learning, which includes the intersection between Reinforcement Learning, Imitation Learning and Representation Learning. My research interest lies in online learning for robotics. I’m interested in demo-driven reinforcement learning combined with pretraining and inductive biases such as 3D vision priors to achieve high sample efficiency. Specifically, I’m interested in learning policies from a limited number of demonstrations and interactions, and continuing learning in an online manner.

Previously, I was a Data Scientist at AWS Professional Services, helping enterprise customers build Machine Learning solutions using AWS.

I also regularly train Olympic Weightlifting, competing in local competitions around the UK.

selected publications

  1. greenaug.gif
    Green Screen Augmentation Enables Scene Generalisation in Robotic Manipulation
    Eugene Teoh, Sumit Patidar, Xiao Ma, and Stephen James