Machine Intelligence Laboratory at Korea University

The Machine Intelligence Laboratory (MiLab) is devoted to the development of computational models and their applications to various researches in the brain and cognitive engineering field. Specifically, we focus on 1) pattern recognition and machine learning algorithms inspired from the neural information processing mechanism in a human brain, 2) brain disease diagnosis or prognosis by analyzing complex patterns inherent in neuroimaging or genetic data, and 3) non-invasive brain-computer interfaces to enhance human performance. [Read more] 

We have developed various machine learning methods, including deep learning for latent high-level feature representations, discriminative sparse regression for feature selection, a Bayesian framework for spatio-spectral filter optimization. The developed methods have been successfully applied to brain disorder diagnosis (Alzheimer's disease and its prodromal stage, mild cognitive impairment), providing insights to the underlying characteristics of structural and functional patterns, and to EEG-based brain-computer interfaces, allowing the investigation of neuro-physiological patterns induced by motor imagery and the prediction of the BCI illiterates. [Learn more]