Improving FIFA Player Agents Decision-Making Architectures Based on Convolutional Neural Networks Through Evolutionary Techniques

Faria, Matheus Prado Prandini; Julia, Rita Maria Silva; Tomaz, Lídia Bononi Paiva

doi:10.1007/978-3-030-61377-8_26

Cited by 3 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A particular class of DNN includes convolutional neural networks (CNN) which are used for image classification, image segmentation, or object detection. They have been gaining in popularity and attention from the scientific community in the recent years as they are at heart of important technological advances such as imitation learning [14] or medical imagery analysis [23] to name a few.…”

Section: Introductionmentioning

confidence: 99%

Use of static surrogates in hyperparameter optimization

Lakhmiri¹,

Digabel²

2021

Preprint

View full text Add to dashboard Cite

Optimizing the hyperparameters and architecture of a neural network is a long yet necessary phase in the development of any new application. This consuming process can benefit from the elaboration of strategies designed to quickly discard low quality configurations and focus on more promising candidates. This work aims at enhancing HyperNOMAD, a library that adapts a direct search derivative-free optimization algorithm to tune both the architecture and the training of a neural network simultaneously, by targeting two keys steps of its execution and exploiting cheap approximations in the form of static surrogates to trigger the early stopping of the evaluation of a configuration and the ranking of pools of candidates. These additions to HyperNOMAD are shown to improve on its resources consumption without harming the quality of the proposed solutions.

show abstract

Section: Introductionmentioning

confidence: 99%