김병형 연구실

Learning non-linear data from the human brain to recognize human affect, analyze spatial-temporal hemispheric structures, and classify physiological data such as EEG, PPG, EMG, and facial expressions.

Developing closed-loop systems that learn emotional reactions to machine behaviors and provide affective feedback to optimize parameters for smooth human-AI interaction and collaboration.

Building interactive AI systems to analyze how contexts of daily life affect emotional changes and establish causal structures between emotions and observable behaviors using wearable technology.

Developing hybrid interfaces with EEG-based classification and eye tracking to enable brain-computer interface control of machine systems like robots and drones in daily life.

Building human-interpretable machine learning systems and evaluating their explanatory efficacy and effects on user mental models, trust, and performance in human-AI interaction.