Detection of Berezinskii-Kosterlitz-Thouless transition via Generative Adversarial Networks

The detection of phase transitions in quantum many-body systems with lowest possible prior knowledge their details is among the most rousing goals flourishing application machine-learning techniques to physical questions. Here, we train a Generative Adversarial Network (GAN) Entanglement Spectrum system bipartition, as extracted by means Matrix Product States ansätze. We are able identify gapless-to-gapped different one-dimensional models looking at machine inability reconstruct outsider data respect training set. foresee that GAN-based methods will become instrumental anomaly schemes applied determination phase-diagrams.