Reinforcement Learning for Adaptive Mesh Refinement

by   Jiachen Yang, et al.

Large-scale finite element simulations of complex physical systems governed by partial differential equations crucially depend on adaptive mesh refinement (AMR) to allocate computational budget to regions where higher resolution is required. Existing scalable AMR methods make heuristic refinement decisions based on instantaneous error estimation and thus do not aim for long-term optimality over an entire simulation. We propose a novel formulation of AMR as a Markov decision process and apply deep reinforcement learning (RL) to train refinement policies directly from simulation. AMR poses a new problem for RL in that both the state dimension and available action set changes at every step, which we solve by proposing new policy architectures with differing generality and inductive bias. The model sizes of these policy architectures are independent of the mesh size and hence scale to arbitrarily large and complex simulations. We demonstrate in comprehensive experiments on static function estimation and the advection of different fields that RL policies can be competitive with a widely-used error estimator and generalize to larger, more complex, and unseen test problems.



There are no comments yet.


page 1

page 5

page 7

page 8

page 13

page 14


A new quality preserving polygonal mesh refinement algorithm for Virtual Element Methods

Mesh adaptivity is a useful tool for efficient solution to partial diffe...

Adaptive Mesh Refinement in Analog Mesh Computers

The call for efficient computer architectures has introduced a variety o...

Control with adaptive Q-learning

This paper evaluates adaptive Q-learning (AQL) and single-partition adap...

Deep Reinforcement Learning with Adjustments

Deep reinforcement learning (RL) algorithms can learn complex policies t...

AMReX: Block-Structured Adaptive Mesh Refinement for Multiphysics Applications

Block-structured adaptive mesh refinement (AMR) provides the basis for t...

Iterative Empirical Game Solving via Single Policy Best Response

Policy-Space Response Oracles (PSRO) is a general algorithmic framework ...

Inf-sup stability implies quasi-orthogonality

We prove new optimality results for adaptive mesh refinement algorithms ...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.