Hong Chul Nam

I am an incoming PhD student at New York University (NYU) under the supervision of Zongyi Li. I completed my Master’s and Bachelor’s at ETH Zurich majoring in electrical engineering. I am currently an AI researcher at Zenithon AI building foundation models for plasma physics. I was a visiting student at Anima’s Lab at Caltech. Previously, I worked in Alsemy, CSEM, Learning & Adaptive Systems Group under Prof. Andreas Krause at ETH Zurich and SAFARI Group under Prof. Onur Mutlu at ETH Zurich, and Korea Military Academy.

My research centers on operator learning and generative models with focus on mathemathical principles and physical correctness. I develop methods that learn maps between function spaces and generative models over functions, such as energy-based operator learning and weakly-supervised training of neural operators from stochastic estimators. I have applied these ideas across scientific and engineering problems, such as solving high-dimensional Poisson equations, unsupervised function-level anomaly detection for erroneous transistor device simulations, and augmenting semiconductor device simulations with generative operators. While interested in proposing new machine learning algorithms, I am equally driven by the scientific applications, with prior experience in modeling transistor reliability and thermal behavior, improving weather forecasting with function-level self-supervised learning, and accelerating virtual page table translation.

Luckily, I have had wide exposure to diverse areas, including (chronological order) neuroscience, robotics, chip design, computer architecture, semiconductor physics, and deep learning. Due to the wide spectrum, I am generally interested in multidisciplinary areas such as AI4Science, scientific foundation models, computational electronics + deep learning, Systems4AI, and AI4EDA.