I am a first-year Ph.D. student in Computer Science at Stanford University. Prior to joining Stanford, I have spent time at NAVER AI and KAIST SILAB, advised by Jinkyoo Park. I hold a Master’s degree in Industrial and Systems Engineering from KAIST and a double Bachelor’s in Control Theory and Automation Engineering from the University of Bologna and Tongji University.
I am originally from Bologna, Italy, but I have had the wonderful opportunity to spend ~20% of my current lifespan in Asia (several years in China, several more in South Korea).
My work spans topics in deep learning, dynamical systems, variational inference and numerical methods. I am broadly interested in ensuring the successes achieved by deep learning methods in computer vision and natural language are extended to other engineering domains.
To achieve this vision, my research follows two fundamental principles:
• High-quality ML software standards common in computer vision and natural language (including model heuristics and massively parallel training) should be exported and adapted to the requirements of other engineering areas.
• Domain expertise in the form of mathematical models and advanced numerical methods should be embedded in the learning model whenever possible, in order to guide other design decisions and ensure a satisfactory baseline performance.
I am a core maintainer and contributor to various open-source libraries for neural differential equations and numerical methods for neural networks.