Abstract
A major challenge ofAI + Science lies in their inherent incompatibility: today’s Al is primarily based on connectionism, while science depends on symbolism. In the first part ofthe talk, I will talk about Kolmogorov-Arnold Networks (KANs) as a solution to synergize both worlds. Inspired by Kolmogorov-Arnold representation theorem, KANs are more aligned with symbolic representations than MLPs, and demonstrate strong accuracy and interpretability. In the second part, I will talk about more broadly the intersection of AI and Science, including science for AI (Poisson Flow Generative Models), science of Al (understanding grokking), and AI for Science (Al scientists).
Speaker
Ziming Liu is a fourth-year PhD student at MIT and IAIFI, working under the supervision of Prof. Max Tegmark. His research lies at the intersection of AI and physics, with a focus on leveraging the synergies between these fields: Physics of AI: Exploring how AI can be as simple as physics. Physics for AI: Developing AI systems that are as natural as physics. AI for Physics: Harnessing the power of AI to advance physics research. Ziming Liu thrives in interdisciplinary collaborations, often working with experts from diverse scientific backgrounds. His research reflects this versatility, encompassing contributions such as Kolmogorov-Arnold networks (Math for AI), Poisson Flow Generative Models (Physics for AI), brain-inspired modular training (Neuroscience for AI), understanding Grokking (Physics of AI), and leveraging conservation laws and symmetries (AI for Physics).