Pehlevan Group

Our goal is to elucidate the theoretical foundations of natural and artificial intelligence through the lens of neural computation. Our research intersects with theoretical and computational neuroscience, deep learning theory, the physics of learning, machine learning, statistical mechanics and high-dimensional statistics. We are particularly interested in how computation emerges in neural networks through the collective dynamics of simple processing units, or neurons. By uncovering these dynamics, we aim to understand the capabilities and limitations of intelligent systems across various contexts, informing future innovations in machine learning and neurotechnology.