Speeding Up Learning Quantum States Through Group Equivariant Convolutional Quantum Ansatze

We develop a theoretical framework for Sn-equivariant convolutional quantum circuits with SU(d) symmetry, building on and significantly generalizing Jordan's permutational quantum computing formalism based on Schur-Weyl duality connecting both SU(d) and Sn actions on qudits. In particular, we utilize the Okounkov-Vershik approach to prove Harrow's statement on the equivalence between SU(d) and Sn irrep bases and to establish the Sn-equivariant convolutional quantum alternating ansatze (Sn-CQA) using Young-Jucys-Murphy elements. We prove that Sn-CQA is able to generate any unitary in any given Sn irrep sector, which may serve as a universal model for a wide array of quantum machine-learning problems with the presence of SU(d) symmetry. Our method provides another way to prove the universality of the quantum approximate optimization algorithm and verifies that four-local SU(d)-symmetric unitaries are sufficient to build generic SU(d)-symmetric quantum circuits up to relative phase factors. We present numerical simulations to showcase the effectiveness of the ansatze to find the ground-state energy of the ./1-./2 antiferromagnetic Heisenberg model on the rectangular and kagome lattices. Our work provides the first application of the celebrated Okounkov-Vershik Sn representation theory to quantum physics and machine learning, from which to propose quantum variational ansatze that strongly suggests to be classically intractable tailored towards a specific optimization problem.