• Home
  • Search
  • Pairwise Gaussian Loss for Convolutional Neural Networks
  • Cite Icon33
  • https://doi.org/10.1109/tii.2019.2963434Copy DOI Icon

Pairwise Gaussian Loss for Convolutional Neural Networks

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Convolutional neural networks (CNNs) have demonstrated great competence in feature representation, and then, achieved a good performance to many classification tasks. Cross-entropy loss, together with softmax, is arguably one of the most commonly used loss functions in CNNs (that is generally called softmax loss). However, the softmax loss can result in a weakly discriminative feature representation since it focuses on the interclass separability rather than the intraclass compactness. This article proposes a pairwise Gaussian loss (PGL) for CNNs that can well address the intraclass compactness through significantly penalizing those similar sample pairs with a relatively large distance. At the same time, PGL can still ensure a good interclass separability. Experiments show that PGL can guarantee that CNNs obtain a better classification performance compared to not only the softmax loss but also others often used in CNNs. Our experiments also show that PGL has a stable convergence for the stochastic gradient descent optimization method in CNNs and a good generalization ability for different structures of CNNs.

Similar Papers
  • Conference Article
  • Citations4

Improving Convolutional Neural Networks Via Compacting Features

  • Apr 01, 2018
  • Liguo Zhou +4
  • Research Article
  • Citations109

A Comprehensive Study on Center Loss for Deep Face Recognition

  • Jan 17, 2019
  • International Journal of Computer Vision
  • Yandong Wen +3
  • Research Article
  • Citations6

More discriminative convolutional neural network with inter-class constraint for classification

  • May 01, 2018
  • Computers & Electrical Engineering
  • Jianchao Fei +4
  • Research Article
  • Citations21

Deep Correlation Feature Learning for Face Verification in the Wild

  • Dec 01, 2017
  • IEEE Signal Processing Letters
  • Weihong Deng +3
  • Research Article
  • Citations83

Gastrointestinal tract classification using improved LSTM based CNN

  • Aug 06, 2020
  • Multimedia Tools and Applications
  • Şaban Öztürk +1
  • Book Chapter
  • Citations2

I-Center Loss for Deep Neural Networks

  • Jan 01, 2018
  • Senlin Cheng +1
  • Book Chapter
  • Citations1

Overall Loss for Deep Neural Networks

  • Jan 01, 2019
  • Hai Huang +2
  • Research Article
  • Citations30

Learning discriminative features via weights-biased softmax loss

  • May 31, 2020
  • Pattern Recognition
  • Xiaobin Li +1
  • Research Article
  • Citations3

An efficient method for determining the optimal convolutional neural network structure based on Taguchi method

  • Oct 07, 2020
  • Journal of Intelligent & Fuzzy Systems
  • Pin-Chan Lee +3
  • PDF
  • Research Article
  • Citations99

Convolutional Neural Network Hyper-Parameters Optimization based on Genetic Algorithms

  • Jan 01, 2018
  • International Journal of Advanced Computer Science and Applications
  • Sehla Loussaief +1
  • Research Article
  • Citations35

Vehicle and Person Re-Identification With Support Neighbor Loss.

  • Feb 01, 2022
  • IEEE transactions on neural networks and learning systems
  • Kai Li +4
  • Supplementary Content
  • Citations9

Convolutional Neural Network based Age Estimation from Facial Image and Depth Prediction from Single Image

  • Jan 01, 2016
  • ANU Open Research (Australian National University)
  • Jiayan Qiu
  • Research Article
  • Citations8

Hatching egg classification based on CNN with channel weighting and joint supervision

  • Oct 23, 2018
  • Multimedia Tools and Applications
  • Lei Geng +5
  • Research Article
  • Citations59

G -Softmax: Improving Intraclass Compactness and Interclass Separability of Features.

  • May 10, 2019
  • IEEE Transactions on Neural Networks and Learning Systems
  • Yan Luo +3
  • Book Chapter
  • Citations6

Streamlining Choice of CNNs and Structure Framing of Convolution Layer

  • Jan 01, 2020
  • Sonika Dahiya +2
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.