Parallel Distributed Implementation of Neuroevolution of Augmenting Topologies in Continuous Control Tasks

This paper proposes a novel distributed implementation of neuroevolution of augmenting topologies method, which, considering the availability of sufficient computational resources, allows drastically speed up the process of optimal neural network configuration search. The proposed solution includes...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2021 IEEE 3rd International Conference on Advanced Trends in Information Theory (ATIT) s. 267 - 271
Hlavní autoři: Achour, Illia, Doroshenko, Anatoliy
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 15.12.2021
Témata:
artificial neural networks
Benchmark testing
cloud computing
Computational efficiency
genetic algorithms
Genomics
NEAT
Network topology
Neural networks
neuroevolution of augmenting topologies
parallel distributed computing
Process control
reinforcement learning
Topology
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This paper proposes a novel distributed implementation of neuroevolution of augmenting topologies method, which, considering the availability of sufficient computational resources, allows drastically speed up the process of optimal neural network configuration search. The proposed solution includes batch genome evaluation for the purpose of performance optimization, fair, and even computational resources usage. The benchmarking shows that the generated neural networks evaluation process can give orders of magnitude increase of efficiency on the demonstrated continuous control task and computational environment.
DOI:10.1109/ATIT54053.2021.9678858