NEURO-LOOP
Brain-Computer Interfaces for Human-in-the-Loop RL
NEURO-LOOP is a reinforcement learning framework that integrates implicit neural feedback into robot policy learning.
Projects
Mapping fNIRS to Agent Performance
Towards Reinforcement Learning from Neural Feedback
This project investigates the relationship between passive neural signals (fNIRS) and artificial agent performance. We analyze correlations between hemodynamic response in the PFC and agent behavior to evaluate whether implicit fNIRS data can act as a feedback signal used to align agent policies with user expectations via RL.
SpheroHunter
Tracking Down Sphero with Occlusions
We design an algorithm that allows the LoCoBot to track and follow a Sphero by predicting its whereabouts despite occlusion in a maze environment.