Abstract:Integrating machine learning techniques in established numerical solvers represents a modern approach to enhancing computational fluid dynamics simulations. Within the lattice Boltzmann method (LBM), the collision operator serves as an ideal entry point to incorporate machine learning techniques to enhance its accuracy and stability. In this work, an invariant neural network is constructed, acting on an equivariant collision operator, optimizing the relaxation rates of non-physical moments. This optimization enhances robustness to symmetry transformations and ensures consistent behavior across geometric operations. The proposed neural collision operator (NCO) is trained using forced isotropic turbulence simulations driven by spectral forcing, ensuring stable turbulence statistics. The desired performance is achieved by minimizing the energy spectrum discrepancy between direct numerical simulations and underresolved simulations over a specified wave number range. The loss function is further extended to tailor numerical dissipation at high wave numbers, ensuring robustness without compromising accuracy at low and intermediate wave numbers. The NCO's performance is demonstrated using three-dimensional Taylor-Green vortex (TGV) flows, where it accurately predicts the dynamics even in highly underresolved simulations. Compared to other LBM models, such as the BGK and KBC operators, the NCO exhibits superior accuracy while maintaining stability. In addition, the operator shows robust performance in alternative configurations, including turbulent three-dimensional cylinder flow. Finally, an alternative training procedure using time-dependent quantities is introduced. It is based on a reduced TGV model along with newly proposed symmetry boundary conditions. The reduction in memory consumption enables training at higher Reynolds numbers, successfully leading to stable yet accurate simulations.

Nebula: Network Enhanced Boltzmann Machine With Universal Local Search Architecture

Restricted Boltzmann machine based algorithm for multi-objective optimization

Superposition Enhanced Nested Sampling

BNEM: A Boltzmann Sampler Based on Bootstrapped Noised Energy Matching

borealis - A generalized global update algorithm for Boolean optimization problems

BEACON: A Bayesian Optimization Strategy for Novelty Search in Expensive Black-Box Systems

AdamMCMC: Combining Metropolis Adjusted Langevin with Momentum-based Optimization

A Unified Particle-Optimization Framework for Scalable Bayesian Sampling

Improving sample efficiency of high dimensional Bayesian optimization with MCMC

Global Neural Network Potential with Explicit Many-Body Functions for Improved Descriptions of Complex Potential Energy Surface.

A Bayesian Optimization through Sequential Monte Carlo and Statistical Physics-Inspired Techniques

Nearest-Neighbours Neural Network architecture for efficient sampling of statistical physics models

Sparse Autoregressive Neural Networks for Classical Spin Systems

Non-convex Bayesian Learning via Stochastic Gradient Markov Chain Monte Carlo

A Probabilistic Niching Evolutionary Computation Framework Based on Binary Space Partitioning

Neural Network and Nearest Neighbor Algorithms for Enhancing Sampling of Molecular Dynamics

Machine Learning Enhanced Collision Operator for the Lattice Boltzmann Method Based on Invariant Networks

Solving Boltzmann Optimization Problems with Deep Learning

Deterministic global optimization with a neighbourhood determination algorithm based on neural networks

NEBULA: Neural Empirical Bayes Under Latent Representations for Efficient and Controllable Design of Molecular Libraries

Fair Sampling of Ground-State Configurations of Binary Optimization Problems.