Quantum phase recognition using quantum tensor networks

Machine learning (ML) has recently facilitated many advances in solving problems related to many-body physical systems. Given the intrinsic quantum nature of these problems, it is natural speculate that quantum-enhanced machine will enable us unveil even greater details than we currently have. With this motivation, paper examines a approach based on shallow variational ansatz inspired by tensor networks for supervised tasks. In particular, first look at standard image classification tasks using Fashion-MNIST dataset and study effect repeating network layers ansatz’s expressibility performance. Finally, use strategy tackle problem phase recognition transverse-field Ising Heisenberg spin models one two dimensions, where were able reach $$\ge 98\%$$ test-set accuracies with both multi-scale entanglement renormalization (MERA) tree (TTN) parametrized circuits.