38 lines
2.3 KiB
Plaintext
38 lines
2.3 KiB
Plaintext
Abstract
|
|
Neuroevolution is a subfield of artificial intelligence that
|
|
leverages evolutionary algorithms to generate and optimize
|
|
artificial neural networks. This technique has proven to be
|
|
successful in solving a wide range of complex problems
|
|
across various domains. The NeuroEvolution of Augmenting
|
|
Topologies (NEAT) is one of the most renowned algorithms
|
|
in neuroevolution. Characterized by its openendedness, it
|
|
starts with minimal networks and progressively evolves both
|
|
the topology and the weights of these networks to optimize
|
|
performance. However, the acceleration techniques employed
|
|
in prevailing NEAT implementations typically rely on par-
|
|
allelism on CPUs, failing to harness the rapidly expand-
|
|
ing computational resources of today. To bridge this gap,
|
|
we present NEATAX, an innovative framework that adapts
|
|
NEAT for execution on hardware accelerators. Built on top
|
|
of the JAX, NEATAX represents networks with varying topo-
|
|
logical structures as tensors with the common shape, facili-
|
|
tating efficient parallel computation using function vectoriza-
|
|
tion. Upon rigorous testing across various tasks, we found
|
|
that NEATAX has the capacity to shrink the computa-
|
|
tion time from hours or even days down to a matter of
|
|
minutes. These results demonstrate the potential of NEATAX
|
|
as a scalable and efficient solution for neuroevolution tasks,
|
|
paving the way for the future application of NEAT in more
|
|
complex and demanding scenarios. NEATAX is available at
|
|
https://github.com/WLS2002/neatax
|
|
|
|
\section{Introduction}
|
|
|
|
Inspired by the principles of natural selection and genetic inheritance, Evolutionary Computation (EC) has emerged
|
|
as a powerful approach in the field of Artificial Intelligence (AI). EC exhibits a robust ability to explore vast
|
|
and complex solution spaces, which is particularly critical when tackling ``black box" optimization problems where the
|
|
internal structure isn't fully visible or understood. Leveraging the power of population-based search, EC navigates
|
|
these complexities to arrive at near-optimal solutions \cite{eiben2015introduction}. However, despite these strengths, recent scholarship has
|
|
highlighted limitations of EC. Important aspects such as ``openendedness" and ``genotype-to-phenotype mappings" warrant
|
|
further attention, especially in light of EC's tendency to rely on small populations and strong selection pressure \cite{miikkulainen_biological_2021}.
|