We will apply computational neurosciences and AI techniques to the diagnosis and treatment of neuropsychiatric disorders such as developmental disorders, mood disorders, schizophrenia, and pain. Specifically, we will contribute to the development and application of (1) biomarkers using machine learning algorithms that can learn from a small number of samples, and (2) advanced neurofeedback based on data-driven selection of intervention targets.

In this research project, we explore technological backgrounds of mathematical models related to information processing in the brain as well as next-generation AI based on such models, review the possibility of realization of such next-generation AI technology and trends of state-of-the-art research, and propose important tasks necessary for the realization of next-generation AI that learns from the dynamic brain particularly from the viewpoint of nonlinear dynamics with a focus on both functions of the normal brain and dysfunctions of the impaired brain. Then, we develop basic mathematical technology for innovative brain-type algorithms, and consider its applicability to robotics and mental illnesses.