Publication:
Learning channel-wise ordered aggregations in deep neural networks

Date

2021

Director

Publisher

Springer
Acceso abierto / Sarbide irekia
Contribución a congreso / Biltzarrerako ekarpena
Versión aceptada / Onetsi den bertsioa

Project identifier

Abstract

One of the most common techniques for approaching image classification problems are Deep Neural Networks. These systems are capable of classifying images with different levels of detail at different levels of detail, with an accuracy that sometimes can surpass even manual classification by humans. Most common architectures for Deep Neural Networks are based on convolutional layers, which perform at the same time a convolution on each input channel and a linear aggregation on the convoluted channels. In this work, we develop a new method for augmenting the information of a layer inside a Deep Neural Network using channel-wise ordered aggregations. We develop a new layer that can be placed at different points inside a Deep Neural Network. This layer takes the feature maps of the previous layer and adds new feature maps by applying several channel-wise ordered aggregations based on learned weighting vectors. We perform several experiments introducing this layer in a VGG neural network and study the impact of the new layer, obtaining better accuracy scores over a sample dataset based on ImageNet. We also study the convergence and evolution of the weighting vectors of the new layers over the learning process, which gives a better understanding of the way the system is exploiting the additional information to gain new knowledge.

Description

Trabajo presentado a la International Conference on Intelligent and Fuzzy Systems, INFUS 2020, 21-23 July 2020, Istanbul, Turkey

Keywords

Neural nets, RNN, Deep Learning, Ordered aggregations

Department

Institute of Smart Cities - ISC

Faculty/School

Degree

Doctorate program

item.page.cita

item.page.rights

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021.

Los documentos de Academica-e están protegidos por derechos de autor con todos los derechos reservados, a no ser que se indique lo contrario.