Replacing neural networks by optimal analytical predictors for the detection of phase transitions

Abstract

Identifying phase transitions and classifying phases of matter is central to understanding the properties and behavior of a broad range of material systems. In recent years, machine-learning (ML) techniques have been successfully applied to perform such tasks in a data-driven manner. However, the success of this approach notwithstanding, we still lack a clear understanding of ML methods for detecting phase transitions, particularly of those that utilize neural networks (NNs). In this work, we derive analytical expressions for the optimal output of three widely used NN-based methods for detecting phase transitions. These optimal predictions correspond to the results obtained in the limit of high model capacity. Therefore, in practice they can, for example, be recovered using sufficiently large, well-trained NNs. The inner workings of the considered methods are revealed through the explicit dependence of the optimal output on the input data. By evaluating the analytical expressions, we can identify phase transitions directly from experimentally accessible data without training NNs, which makes this procedure favorable in terms of computation time. Our theoretical results are supported by extensive numerical simulations covering, e.g., topological, quantum, and many-body localization phase transitions. We expect similar analyses to provide a deeper understanding of other classification tasks in condensed matter physics.

19+20 pages, 8+12 figures; improved presentation and numerical implementation, added discussion on model capacity and data set size

References in corpus (11)

• Probing many-body dynamics on a 51-atom quantum simulator
• Quantum Simulation of Antiferromagnetic Spin Chains in an Optical Lattice
• Learning phase transitions by confusion
• Quantum advantage in learning from experiments
• QuSpin: a Python Package for Dynamics and Exact Diagonalisation of Quantum Many Body Systems part I: spin chains
• Machine learning vortices at the Kosterlitz-Thouless transition
• Entanglement and topological entropy of the toric code at finite temperature
• Machine Learning of Explicit Order Parameters: From the Ising Model to SU(2) Lattice Gauge Theory
• Deep Learning the Quantum Phase Transitions in Random Electron Systems: Applications to Three Dimensions
• Direct Evidence of the Discontinuous Character of the Kosterlitz-Thouless Jump
• Detecting ergodic bubbles at the crossover to many-body localization using neural networks

Cited by in corpus (4)

• A learning by confusion approach to characterize phase transitions
• Learning phase transitions from regression uncertainty: A new regression-based machine learning approach for automated detection of phases of matter
• Efficient and quantum-adaptive machine learning with fermion neural networks
• Snake net and balloon force with a neural network for detecting multiple phases