Flow Matching at Scale: A Machine Learning Framework for Efficient Large-Size Sampling of Many-Body Systems

• URL: http://arxiv.org/abs/2508.15318v3
• Date: Fri, 05 Sep 2025 04:25:10 GMT
• Title: Flow Matching at Scale: A Machine Learning Framework for Efficient Large-Size Sampling of Many-Body Systems
• Abstract summary: We propose a machine learning framework based on Flow Matching to overcome the scaling limitations of Markov Chain Monte Carlo methods.<n>We demonstrate its capability in the 2D XY model, where a single network, trained only on configurations from a small lattice at sparse temperature points, generates reliable samples for a significantly larger system.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We propose a machine learning framework based on Flow Matching to overcome the scaling limitations of Markov Chain Monte Carlo (MCMC) methods. We demonstrate its capability in the 2D XY model, where a single network, trained only on configurations from a small ($32\times 32$) lattice at sparse temperature points, generates reliable samples for a significantly larger system ($128\times 128$) across a continuous temperature range without retraining. The generated configurations show strong agreement with key thermodynamic observables and correctly capture the signatures of the Berezinskii-Kosterlitz-Thouless (BKT) transition. This dual generalization is enabled by the Flow Matching framework, which allows us to learn a continuous, temperature-conditioned mapping. At the same time, the inductive biases of the underlying CNN architecture ensure that the learned local physical rules are scale-invariant. This "train-small, generate-large" capability offers a powerful and efficient alternative for studying critical phenomena. The method can be directly applied to other classical or quantum many-body systems described by continuous fields on a lattice. Furthermore, this framework can serve as a powerful proposal generator in a hybrid scheme with MCMC, dramatically accelerating high-precision studies of the thermodynamic limit.

Related papers

• Generative Nested Sampling of Atomistic Thermodynamic Landscapes [37.547720439639015]
  Flow-based nested sampling has removed the bottleneck for gravitational-wave (GW) inference.<n>We show that LJ disks in PBC reduce energy evaluations by over two orders of magnitude and wall-clock time by roughly one third.<n>For LJ disks in PBC, the algorithm reduces energy evaluations by over two orders of magnitude and wall-clock time by roughly one third.
  arXiv  Detail & Related papers  (2026-09-02T22:21:10Z)
• Diffusion-warm sampling of the XY model enables fast thermalization at scale [1.9116784879310027]
  We introduce a novel technique for scalable sampling of spin-system states with continuous symmetries using diffusion models.<n>We show that our technique addresses the shortfalls of the Markov chain Monte Carlo in generalization to varying system sizes.
  arXiv  Detail & Related papers  (2026-06-29T18:06:09Z)
• Statistical Channel Fingerprint Construction for Massive MIMO: A Unified Tensor Learning Framework [51.39651336043647]
  Channel fingerprint (CF) is considered a key enabler for facilitating the acquisition of channel state information inIMO communication systems.<n>We propose a novel type of CF that stores statistical CSI (sCSI) at each potential location, referred to as statistical CF (sCF)
  arXiv  Detail & Related papers  (2026-04-30T08:28:45Z)
• Scaling Autoregressive Models for Lattice Thermodynamics [5.904448014233719]
  Markov-chain Monte Carlo sampling suffers from slow convergence and critical slowing down near phase transitions.<n>Existing autoregressive models (ARMs) generate configurations in a fixed sequential order and incur high memory and training costs.<n>Here, we develop a framework combining any-order ARMs, which generate configurations flexibly by conditioning on any known subset of lattice sites, with models (MAMs) which approximate the probability of any partial configuration in a single forward pass.
  arXiv  Detail & Related papers  (2026-03-16T01:03:59Z)
• Lowering the temperature of two-dimensional fermionic tensor networks with cluster expansions [0.0]
  Suzuki-Trotter decomposition is the main technique for representing a Hamiltonian composed of strictly short-ranged interactions.<n>We extend the cluster expansion to two-dimensional fermionic systems, and employ it to construct projected entangled-pair operator (PEPO) approximations of Gibbs states.
  arXiv  Detail & Related papers  (2026-02-25T16:58:58Z)
• Capturing the Topological Phase Transition and Thermodynamics of the 2D XY Model via Manifold-Aware Score-Based Generative Modeling [0.0]
  We propose a new Score-Based Generative Modeling framework for continuous spin systems.<n>We show that our method estimates the theoretical Boltzmann score with superior precision compared to standard diffusion models.<n>We demonstrate zero-shot generalization to unseen lattice sizes, accurately recovering the physics of variable system scales without retraining.
  arXiv  Detail & Related papers  (2026-02-07T13:50:20Z)
• Approaching the Thermodynamic Limit with Neural-Network Quantum States [0.6999740786886536]
  We introduce the Spatial Attention mechanism, a minimal and physically interpretable inductive bias for Neural-Network Quantum States.<n>This bias stabilizes large-scale optimization and enables access to thermodynamic-limit physics.<n>We show that our approach achieves a state-of-the-art energies for a $20times20$ square lattice, outperforming Residual Convolutional Neural Networks.
  arXiv  Detail & Related papers  (2026-02-02T19:00:05Z)
• Optimized Machine Learning Methods for Studying the Thermodynamic Behavior of Complex Spin Systems [0.0]
  convolutional neural networks (CNNs) are an efficient and versatile tool for the analysis of critical and low-temperature phase states in spin system models.<n>It is shown that CNNs substantially reduce the root-mean-square error (RMSE) compared with fully connected architectures.
  arXiv  Detail & Related papers  (2025-12-08T11:33:24Z)
• Towards Efficient General Feature Prediction in Masked Skeleton Modeling [59.46799426434277]
  We propose a novel General Feature Prediction framework (GFP) for efficient mask skeleton modeling.<n>Our key innovation is replacing conventional low-level reconstruction with high-level feature prediction that spans from local motion patterns to global semantic representations.
  arXiv  Detail & Related papers  (2025-09-03T18:05:02Z)
• Learning Kinetic Monte Carlo stochastic dynamics with Deep Generative Adversarial Networks [0.0]
  We show that Generative Adversarial Networks (GANs) may be fruitfully exploited to learn dynamics.<n>We showcase the application to a two-dimensional, many-particle system, focusing on surface-step fluctuations and on the related time-dependent roughness.
  arXiv  Detail & Related papers  (2025-07-29T12:48:03Z)
• Leveraging recurrence in neural network wavefunctions for large-scale simulations of Heisenberg antiferromagnets: the square lattice [1.9681634372790209]
  Machine-learning-based variational Monte Carlo simulations are a promising approach for targeting quantum many-body ground states.<n>We employ recurrent neural networks (RNNs) as a variational ans"atze, and leverage their recurrent nature to simulate the ground states of progressively larger systems.<n>We show that we are able to systematically improve the accuracy of the results from our simulations by increasing the training time.
  arXiv  Detail & Related papers  (2025-02-24T13:35:23Z)
• Language Models as Zero-shot Lossless Gradient Compressors: Towards General Neural Parameter Prior Models [56.00251589760559]
  Large language models (LLMs) can act as gradient priors in a zero-shot setting.<n>We introduce LM-GC, a novel method that integrates LLMs with arithmetic coding.<n>Experiments indicate that LM-GC surpasses existing state-of-the-art lossless compression methods.
  arXiv  Detail & Related papers  (2024-09-26T13:38:33Z)
• Enhancing lattice kinetic schemes for fluid dynamics with Lattice-Equivariant Neural Networks [79.16635054977068]
  We present a new class of equivariant neural networks, dubbed Lattice-Equivariant Neural Networks (LENNs) Our approach develops within a recently introduced framework aimed at learning neural network-based surrogate models Lattice Boltzmann collision operators. Our work opens towards practical utilization of machine learning-augmented Lattice Boltzmann CFD in real-world simulations.
  arXiv  Detail & Related papers  (2024-05-22T17:23:15Z)
• CWF: Consolidating Weak Features in High-quality Mesh Simplification [50.634070540791555]
  We propose a smooth functional that simultaneously considers all of these requirements. The functional comprises a normal anisotropy term and a Centroidal Voronoi Tessellation (CVT) energy term.
  arXiv  Detail & Related papers  (2024-04-24T05:37:17Z)
• Message-Passing Neural Quantum States for the Homogeneous Electron Gas [41.94295877935867]
  We introduce a message-passing-neural-network-based wave function Ansatz to simulate extended, strongly interacting fermions in continuous space. We demonstrate its accuracy by simulating the ground state of the homogeneous electron gas in three spatial dimensions.
  arXiv  Detail & Related papers  (2023-05-12T04:12:04Z)
• A framework for efficient ab initio electronic structure with Gaussian Process States [0.0]
  We present a framework for the efficient simulation of realistic fermionic systems with modern machine learning inspired representations of quantum many-body states. We show competitive accuracy for systems with up to 64 electrons, including a simplified (yet fully ab initio) model of the Mott transition in three-dimensional hydrogen.
  arXiv  Detail & Related papers  (2023-02-02T13:40:38Z)
• Quantum Markov Chain Monte Carlo with Digital Dissipative Dynamics on Quantum Computers [52.77024349608834]
  We develop a digital quantum algorithm that simulates interaction with an environment using a small number of ancilla qubits. We evaluate the algorithm by simulating thermal states of the transverse Ising model.
  arXiv  Detail & Related papers  (2021-03-04T18:21:00Z)
• Machine learning for rapid discovery of laminar flow channel wall modifications that enhance heat transfer [56.34005280792013]
  We present a combination of accurate numerical simulations of arbitrary, flat, and non-flat channels and machine learning models predicting drag coefficient and Stanton number. We show that convolutional neural networks (CNN) can accurately predict the target properties at a fraction of the time of numerical simulations.
  arXiv  Detail & Related papers  (2021-01-19T16:14:02Z)
• Kernel and Rich Regimes in Overparametrized Models [69.40899443842443]
  We show that gradient descent on overparametrized multilayer networks can induce rich implicit biases that are not RKHS norms. We also demonstrate this transition empirically for more complex matrix factorization models and multilayer non-linear networks.
  arXiv  Detail & Related papers  (2020-02-20T15:43:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.