Hidenori Tanaka is a physicist and AI researcher whose work focuses on understanding the fundamental mechanisms of deep learning and intelligence through the lens of physics. He is Group Leader of the Physics of AI Group at NTT Research and leads the CBS-NTT Program in Physics of Intelligence at Harvard University. He is also an Affiliate Member of MIT’s Institute for AI and Fundamental Interactions (IAIFI).
Born and raised in Tokyo, Japan, he received his B.S. in Physics from Kyoto University and his Ph.D. in Applied Physics from Harvard University, with a minor in Computational Science & Engineering. He later held a Masason Postdoctoral Fellowship at Stanford University.
His research sits at the intersection of physics, machine learning, neuroscience, and psychology. He studies how AI systems learn, generalize, and develop emergent abilities, with the goal of building more trustworthy, interpretable, and energy-efficient AI. His papers have appeared in venues including NeurIPS, ICLR, ICML, Neuron, PNAS, and Physical Review Letters.
His professional interests include AI alignment and trustworthiness, learning dynamics, mechanistic understanding of neural networks, AI efficiency and sustainability, and links between biological and artificial intelligence. He is particularly interested in interdisciplinary collaborations connecting physics, neuroscience, psychology, and computer science, including applications in education and psychiatry.
