首页|Study Findings on Machine Learning Are Outlined in Reports fromFederal Universi ty Rio Grande do Sul (Enhancing AutoencoderbasedMachine Learning Through the U se of Process Control Gainand Relative Gain Arrays As Cost Functions)
Study Findings on Machine Learning Are Outlined in Reports fromFederal Universi ty Rio Grande do Sul (Enhancing AutoencoderbasedMachine Learning Through the U se of Process Control Gainand Relative Gain Arrays As Cost Functions)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning DailyNews Daily News – Current study results on Machine Learn ing have been published. According to newsreporting from Porto Alegre, Brazil, by NewsRx journalists, research stated, “Autoencoders are neuralnetworks utiliz ed for unsupervised learning and reconstructing input data, making them helpful in foranalyzing industrial process data. To enhance their effectiveness, we int roduce two cost functions basedon the Gain Matrix and Relative Gain Array (RGA) concepts, referred to in this paper as Gain Autoencoder(GAE) and Relative Gain Autoencoder (RGAE).”Financial support for this research came from Coordenacao de Aperfeicoamento de Pessoal de NivelSuperior (CAPES).The news correspondents obtained a quote from the research from Federal Universi ty Rio Grandedo Sul, “These cost functions aid in reducing dimensionality and i mproving the model’s performance inindustrial settings. This article delves int o applying of these functions in machine learning, particularly inautoencoders, to predict Mooney viscosity in styrene butadiene rubber (SBR) production. The f indingsindicate that the proposed GAE and RGAE models outperform traditional li near (linear regression) andnonlinear models (SVR), as evidenced by an increase d explained variance of up to 10% and a decrease inmean square er ror of up to 13%.”
Porto AlegreBrazilSouth AmericaCyb orgsEmergingTechnologiesMachine LearningFederal University Rio Grande do Sul
NETL
NSTL
