Sampling the XY model with an attention-based neural network
Presents novel ML architecture for statistical physics sampling with demonstrated performance improvements
AI Summary
Researchers developed a -based neural network that outperforms existing samplers for the XY model lattice system using physics-informed attention mechanisms.
Excerpt
We present a transformer-based variational autoregressive network for sampling lattice systems with continuous degrees of freedom and apply it to the two-dimensional XY model near the Berezinskii-Kosterlitz-Thouless transition. The architecture is combined with physics-informed ingredients: locality-informed attention and approximate probabilities. The resulting sampler significantly outperforms existing neural samplers. We train the model on lattices up to $96\times96$, verify with a mode-dropp