Mutual information of spin systems from autoregressive neural networks
Preprint
Abstract
We describe a direct approach to estimate bipartite mutual information of a classical spin system based on Monte Carlo sampling enhanced by autoregressive neural networks. It allows studying arbitrary geometries of subsystems and can be generalized to classical field theories. We demonstrate it on the Ising model for four partitionings, including a multiply-connected even-odd division. We show that the area law is satisfied for temperatures away from the critical temperature: the constant term is universal, whereas the proportionality coefficient is different for the even-odd partitioning.Keywords
All Related Versions
This publication has 27 references indexed in Scilit:
- Continuous-Mixture Autoregressive Networks Learning the Kosterlitz-Thouless TransitionChinese Physics Letters, 2022
- Hierarchical autoregressive neural networks for statistical systemsComputer Physics Communications, 2022
- Quantum Information Meets Quantum Matter2019
- Detecting classical phase transitions with Renyi mutual informationPhysical Review B, 2013
- Quantum Monte Carlo calculation of entanglement Rényi entropies for generic quantum systemsPhysical Review B, 2012
- Mutual information in classical spin modelsJournal of Statistical Mechanics: Theory and Experiment, 2011
- Shannon and entanglement entropies of one- and two-dimensional critical wave functionsPhysical Review B, 2009
- The Large-N Limit of Superconformal Field Theories and SupergravityInternational Journal of Theoretical Physics, 1999
- Metropolized independent sampling with comparisons to rejection sampling and importance samplingStatistics and Computing, 1996
- On geometric entropyPhysics Letters B, 1994