معرفی
Hikaru Ibayashi is a Researcher at Meta's Instagram organization. He holds a Ph.D. in Computer Science from the University of Southern California and dual degrees in Computer Science (M.S.) and Physics (B.S.) from the University of Tokyo. His research focuses on deep neural networks, physics-informed machine learning, and optimization algorithms.
- Ph.D.: University of Southern California (Computer Science)
- M.S.: University of Tokyo (Computer Science)
- B.S.: University of Tokyo (Physics)
Research interests include scalable neural network-based quantum molecular dynamics simulations, analysis of SGD behavior in non-convex landscapes, and adaptive fluid simulation techniques. His work bridges machine learning theory with computational physics and graphics applications.
Key contributions include the Allegro-Legato model for robust molecular dynamics simulations and novel approaches for liquid simulation using dynamically warping grids. Notable collaborations involve institutions like the University of Tokyo, IST Austria, and TU München.
