Particle Monte Carlo methods for Lattice Field Theory
This work provides a strong classical baseline for machine learning assisted sampling in lattice field theory, raising the bar for when learned proposals are justified, which is important for researchers in computational physics and machine learning.
The paper tackled high-dimensional multimodal sampling problems in lattice field theory by showing that GPU-accelerated particle Monte Carlo methods, such as Sequential Monte Carlo and nested sampling, match or outperform state-of-the-art neural samplers in sample quality and wall-clock time on standard benchmarks, while also estimating the partition function.
High-dimensional multimodal sampling problems from lattice field theory (LFT) have become important benchmarks for machine learning assisted sampling methods. We show that GPU-accelerated particle methods, Sequential Monte Carlo (SMC) and nested sampling, provide a strong classical baseline that matches or outperforms state-of-the-art neural samplers in sample quality and wall-clock time on standard scalar field theory benchmarks, while also estimating the partition function. Using only a single data-driven covariance for tuning, these methods achieve competitive performance without problem-specific structure, raising the bar for when learned proposals justify their training cost.