New RLNF System Trains AI to Decode Human Preferences From Brain Signals

Rahul Chhabra developed RL with Neural Feedback (RLNF), a reinforcement learning technique that enables AI models to optimize outputs based on neural activity. The system was demonstrated by morphing faces in real time to match viewer preferences decoded from brain signals. RLNF bypasses traditional feedback methods like clicks or ratings to gather preference data directly from the human brain.