Strongly interacting electrons exhibit a rich variety of emergent phenomena in quantum materials, from the fractional quantum Hall effect to chiral superconductivity. Yet our understanding of these systems remains limited by the difficulty of solving the many-electron Schrödinger equation in a vast Hilbert space.
I will present a “first-principles AI” framework that uses neural networks as universal and systematically improvable ansätze for many-electron quantum states. Crucially, these neural wavefunctions are optimized entirely by energy minimization, without any external training data or hand-engineered physics knowledge. These architectures are provably universal approximators of fermionic wavefunctions while enforcing antisymmetry, providing a rigorous foundation for tackling the many-electron problem with AI.
Within this framework, I will highlight two new findings. First, I will show how first-principles AI discovers an electron quasicrystal phase in a single semiconductor quantum well, where interacting electrons spontaneously form a twisted-bilayer-like structure with quasiperiodic charge order. Second, I will demonstrate the crystallization of a fractional quantum Hall liquid at strong Landau-level mixing, a regime that has remained inaccessible to previous methods.