A physics-inspired transformer with sparse attention and FlashAttention enables up to 100x faster sampling of large spin-glass systems, providing distributions, free energies, and overlaps for SK and EA models where prior ML methods fail at some temperatures.
Batchtnmc: Efficient sampling of two- dimensional spin glasses using tensor network monte carlo
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cond-mat.dis-nn 2representative citing papers
Global Annealing Monte Carlo with ML global moves plus local updates outperforms Simulated Annealing and is more robust than Population Annealing on 3D Ising spin glasses without hyperparameter tuning.
citing papers explorer
-
Scalable Physics-Inspired Transformers for Spin Glasses
A physics-inspired transformer with sparse attention and FlashAttention enables up to 100x faster sampling of large spin-glass systems, providing distributions, free energies, and overlaps for SK and EA models where prior ML methods fail at some temperatures.