Ya-chao Fan

About

Ya-chao Fan is a leading researcher at the intersection of neuroscience, artificial intelligence, and brain-computer interfaces (BCIs). Their primary work focuses on integrating human neural signals with machine learning, particularly using electroencephalography (EEG) to enhance reinforcement learning (RL) systems. Fan’s most cited study, "Deep reinforcement learning from error-related potentials via an EEG-based brain-computer interface" (2018, 28 citations), tackles a critical bottleneck in deep RL: the inability to learn from human feedback in real-time, real-world settings. By decoding error-related potentials (ErrPs) from EEG data, Fan demonstrated that a BCI could directly communicate human evaluative signals to an RL agent, bypassing the delays and constraints of traditional preference-based methods. This breakthrough enables more natural, continuous human-robot interaction, allowing machines to adapt instantly to user intent. Fan’s work is foundational for advancing assistive robotics, autonomous systems, and adaptive AI, bridging the gap between cognitive neuroscience and practical machine learning. Their contributions have been recognized for pushing the boundaries of how humans can intuitively teach and collaborate with intelligent systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Deep reinforcement learning from error-related potentials via an EEG-based brain-computer interface

   28 citations · 2018

Key Collaborators

Contact & Links