홈 › Papers

Multi-Agent Reinforcement Learning for Adaptive Mesh Refinement

2022-11-02 · Jiachen Yang, Ketan Mittal, Tarik Dzanic, Socratis Petrides, Brendan Keith, Brenden Petersen, Daniel Faissol, Robert Anderson

Adaptive mesh refinement (AMR) is necessary for efficient finite element simulations of complex physical phenomenon, as it allocates limited computational budget based on the need for higher or lower resolution, which varies over space and time. We present a novel formulation of AMR as a fully-cooperative Markov game, in which each element is an independent agent who makes refinement and de-refinement choices based on local information. We design a novel deep multi-agent reinforcement learning (MARL) algorithm called Value Decomposition Graph Network (VDGN), which solves the two core challenges that AMR poses for MARL: posthumous credit assignment due to agent creation and deletion, and unstructured observations due to the diversity of mesh geometries. For the first time, we show that MARL enables anticipatory refinement of regions that will encounter complex features at future times, thereby unlocking entirely new regions of the error-cost objective landscape that are inaccessible by traditional methods based on local error estimators. Comprehensive experiments show that VDGN policies significantly outperform error threshold-based policies in global error and cost metrics. We show that learned policies generalize to test problems with physical features, mesh geometries, and longer simulation times that were not seen in training. We also extend VDGN with multi-objective optimization capabilities to find the Pareto front of the tradeoff between cost and error.

📄 PDF Abstract BibTeX arXiv:2211.00801

Code (1)

011235813/marl-amr 공식 구현 tf

Tasks

Multi-agent Reinforcement Learning reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Methods 이 논문이 사용한 방법론

Test 설명 없음

Swarm Reinforcement Learning For Adaptive Mesh Refinement

2023-04-03 · NeurIPS 2023 11 · Niklas Freymuth, Philipp Dahlinger, Tobias Würth, Simon Reisch 외

Adaptive Mesh Refinement (AMR) enhances the Finite Element Method, an important technique for simulating complex problems in engineering, by dynamically refining mesh regions, enabling a favorable trade-off between compu…

reinforcement-learning Reinforcement Learning

Adaptive Swarm Mesh Refinement using Deep Reinforcement Learning with Local Rewards

2024-06-12 · Niklas Freymuth, Philipp Dahlinger, Tobias Würth, Simon Reisch 외

Simulating physical systems is essential in engineering, but analytical solutions are limited to straightforward problems. Consequently, numerical methods like the Finite Element Method (FEM) are widely used. However, th…

Deep Reinforcement Learning

Deep Reinforcement Learning for Adaptive Mesh Refinement

2022-09-25 · Corbin Foucart, Aaron Charous, Pierre F. J. Lermusiaux

Finite element discretizations of problems in computational physics often rely on adaptive mesh refinement (AMR) to preferentially resolve regions containing important features during simulation. However, these spatial r…

Decision Making Deep Reinforcement Learning reinforcement-learning Reinforcement Learning+2

Reinforcement Learning for Adaptive Mesh Refinement

2021-03-01 · Jiachen Yang, Tarik Dzanic, Brenden Petersen, Jun Kudo 외

Large-scale finite element simulations of complex physical systems governed by partial differential equations (PDE) crucially depend on adaptive mesh refinement (AMR) to allocate computational budget to regions where hig…

Deep Reinforcement Learning Inductive Bias reinforcement-learning Reinforcement Learning+1

Quasi-optimal $hp$-finite element refinements towards singularities via deep neural network prediction

2022-09-13 · Tomasz Sluzalec, Rafal Grzeszczuk, Sergio Rojas, Witold Dzwinel 외

We show how to construct the deep neural network (DNN) expert to predict quasi-optimal $hp$-refinements for a given computational problem. The main idea is to train the DNN expert during executing the self-adaptive $hp$-…