INGLÉS EPUB con DRM eBook

ENHANCING DEEP LEARNING PERFORMANCE USING DISPLACED RECTIFIER LINEAR UNIT | eBook

Por

DAVID MACÊDO

(0)
La compra de eBooks para España debe realizarse desde: vivlio.casadellibro.com

Sinopsis

Recently, deep learning has caused a significant impact on computer vision, speech recognition, and natural language understanding. In spite of the remarkable advances, deep learning recent performance gains have been modest and usually rely on increasing the depth of the models, which often requires more computational resources such as processing time and memory usage. To tackle this problem, we turned our attention to the interworking between the activation functions and the batch normalization, which is virtually mandatory currently. In this work, we propose the activation function Displaced Rectifier Linear Unit (DReLU) by conjecturing that extending the identity function of ReLU to the third quadrant enhances compatibility with batch normalization. Moreover, we used statistical tests to compare the impact of using distinct activation functions (ReLU, LReLU, PReLU, ELU, and DReLU) on the learning speed and test accuracy performance of VGG and Residual Networks state-of-the-art models. These convolutional neural networks were trained on CIFAR-10 and CIFAR-100, the most commonly used deep learning computer vision datasets. The results showed DReLU speeded up learning in all models and datasets. Besides, statistical significant performance assessments (p

Léelo en cualquier dispositivo


Ficha Técnica

Editorial: Editora Dialética

ISBN: 9786525230757

Idioma: Inglés

Fecha de lanzamiento: 25/03/2022

Especificaciones del producto

APP TAGUS

App gratuita de lectura Tagus

Ahora tu lectura es multidispositivo, con la App de Tagus, puedes tener todos tus libros en tu tablet y smartphone. Aprovecha cualquier momento para seguir disfrutando de las lecturas que más te gustan.

Ver más

Reseñas sobre ENHANCING DEEP LEARNING PERFORMANCE USING DISPLACED RECTIFIER LINEAR UNIT (EBOOK)

Comparte tu experiencia con la comunidad lectora.

0

0 Reseñas

5 0
4 0
3 0
2 0
1 0

Sólo por opinar entras en el sorteo mensual de tres tarjetas regalo valoradas en
20€

Los eBooks más vendidos de la semana