I am interested in modeling the brain in silico. In micro scale I’m interested in the learning and memory of the neural network, as to how the information is stored within a deep learning model or how the neural architectures affect learning. In macro scale I’m interested in how the deep learning modules could act like brain regions which interact in synchrony to perform cognitive tasks.

I am currently investigating various cognitive modules of reinforcement learning such as the “model of self”, figuring out whether they emerge naturally and why the modules are needed. Previously I researched in Seoul National University Hospital, South Korea, developing continual learning algorithms, privacy preserving deep learning framework for hospitals, and medical deep learning applications. Before then, I researched voice conversion using CycleGAN as an undergraduate research assistant.

As many characters in myths and legends are humanoids, I believe we humans create what assimilates us, and we could reflect on ourselves from our creations. My preferred example of this reflective creation is the computer design. As the computer was originally designed to simulate the human’s thought process rapidly, I am fascinated by how the computer operates similarly to how humans think, and believe that there lies much potential cognitive insights within the computer. Comparing the mechanisms and failures between the computer and the brain is one of my methods to gain insights for modeling the cognitive functions in silico.

I am warm-hearted and optimistic. Feel free to email if you want to know more about me or my research.