Opinion-guided reinforcement learning (pre-print)

The full pre-print of our work on Opinion-guided reinforcement learning with Kyanna Dagenais is now available on arXiv: https://arxiv.org/abs/2405.17287. Abstract. Human guidance is often desired in reinforcement learning to improve …

Kyanna Dagenais joins lab

Kyanna Dagenais joined our lab as an undergraduate research assistant, working on the topic of uncertainty in reinforcement learning. Kyanna’s work will help foster more performant and (resource-)efficient machine learning …