Authors:
Matheus Faria
;
Etienne Julia
;
Henrique Fernandes
;
Marcelo Zanchetta do Nascimento
and
Rita Julia
Affiliation:
Computer Science Department, Federal University of Uberlândia, Uberlândia, Minas Gerais, Brazil
Keyword(s):
Deep Learning, Convolutional Neural Network, Genetic Algorithm, Bayesian Optimization, Video Games, Game Events, Classification, Dota2.
Abstract:
Game logs are an important part of the player experience analysis in literature. They describe the major actions and events (related to the players or other elements) that affect the progress of a game. In most existing games (especially popular commercial games like FIFA, Dota2 and Valorant), their access is typically restricted to the game’s developers. Deep Learning (DL) approaches have been proposed to perform game event classification from videos. However, retrieving relevant information about these game events (normally associated with actions performed by players) in real-time is still a challenge. Existing approaches require high computational power that serves as an additional issue. In this sense, the present paper investigates a set of approaches that aim to reduce the computational cost of DL-based models - more specifically, Convolutional Neural Networks (CNN) based on Residual Nets architectures - through Genetic Algorithm and Bayesian Optimization. This investigation i
s carried out in the context of Dota2 game event classification. The comparative analysis showed that the models obtained herein achieved a classification performance as good as the models of the state-ofthe-art considering the Dota2 dataset, but with significantly fewer parameters. Thus, this work can help in the generation of optimized CNNs for real-time applications.
(More)