Mapping fNIRS Signals to Agent Performance: Toward Reinforcement Learning from Neural Feedback

Reinforcement Learning from Human Feedback (RLHF) is a methodology that aligns agent behavior with human preferences by integrating human feedback into the agent's training process. We introduce a possible framework that employs passive Brain-Computer Interfaces (BCI) to guide agent training from implicit neural signals. We present and release a novel dataset of functional near-infrared spectroscopy (fNIRS) recordings collected from 25 human participants across three domains: a Pick-and-Place Robot, Lunar Lander, and Flappy Bird. We train classifiers to predict levels of agent performance (optimal, sub-optimal, or worst-case) from windows of preprocessed fNIRS feature vectors, achieving an average F1 score of 67% for binary classification and 46% for multi-class models averaged across conditions and domains. We also train regressors to predict the degree of deviation between an agent's chosen action and a set of near-optimal policies, providing a continuous measure of performance. We evaluate cross-subject generalization and demonstrate that fine-tuning pre-trained models with a small sample of subject-specific data increases average F1 scores by 17% and 41% for binary and multi-class models, respectively. Our work demonstrates that mapping implicit fNIRS signals to agent performance is feasible and can be improved, laying the foundation for future brain-driven RLHF systems.

Mapping Neural Signals to Agent Performance, A Step Towards Reinforcement Learning from Neural Feedback

Implicit Human-in-the-Loop Reinforcement Learning (HITL-RL) is a methodology that integrates passive human feedback into autonomous agent training while minimizing human workload. However, existing methods often rely on active instruction, requiring participants to teach an agent through unnatural expression or gesture. We introduce NEURO-LOOP, an implicit feedback framework that utilizes the intrinsic human reward system to drive human-agent interaction. This work demonstrates the feasibility of a critical first step in the NEURO-LOOP framework: mapping brain signals to agent performance. Using functional near-infrared spectroscopy (fNIRS), we design a dataset to enable future research using passive Brain-Computer Interfaces for Human-in-the-Loop Reinforcement Learning. Participants are instructed to observe or guide a reinforcement learning agent in its environment while signals from the prefrontal cortex are collected. We conclude that a relationship between fNIRS data and agent performance exists using classical machine learning techniques. Finally, we highlight the potential that neural interfaces may offer to future applications of human-agent interaction, assistive AI, and adaptive autonomous systems.