Effective Training of PINNs by Combining CMA-ES with Gradient Descent
Created by W.Langdon from
gp-bibliography.bib Revision:1.7975
- @InProceedings{liu:2024:CEC,
-
author = "Lin Liu and Yuan Yuan",
-
title = "Effective Training of {PINNs} by Combining {CMA-ES}
with Gradient Descent",
-
booktitle = "2024 IEEE Congress on Evolutionary Computation (CEC)",
-
year = "2024",
-
editor = "Bing Xue",
-
address = "Yokohama, Japan",
-
month = "30 " # jun # " - 5 " # jul,
-
publisher = "IEEE",
-
keywords = "genetic algorithms, genetic programming, Training,
Memetics, Neural networks, Evolutionary computation,
Pareto optimization, Prediction algorithms, Linear
programming, Physics-informed neural networks,
evolutionary algorithm, gradient descent, memetic
algorithm, multi-objective optimization",
-
isbn13 = "979-8-3503-0837-2",
-
DOI = "doi:10.1109/CEC60901.2024.10611964",
-
abstract = "Physics-Informed Neural Networks (PINNs) have recently
received increasing attention, however, optimising the
loss function of PINNs is notoriously difficult, where
the landscape of the loss function is often highly
non-convex and rugged. Local optimisation methods based
on gradient information can converge quickly but are
prone to being trapped in local minima for training
PINNs. Evolutionary algorithms (EAs) are well known for
the global search ability, which can help escape from
local minima. It has been reported in the literature
that EAs show some advantages over gradient-based
methods in training PINNs. Inspired by the Memetic
Algorithm, we combine global-search based EAs and
local-search based batch gradient descent in order to
make the best of both word. In addition, since the PINN
loss function is composed of multiple terms, balancing
these terms is also a challenging issue. Therefore, we
also attempt to combine EAs with multiple-gradient
descent algorithm for multi-objective optimisation. Our
experiments provide strong evidence for the superiority
of the above algorithms.",
-
notes = "also known as \cite{10611964}
WCCI 2024",
- }
Genetic Programming entries for
Lin Liu
Yuan Yuan
Citations