Interpolating convolutional neural networks using batch normalization
Perceiving a visual concept as a mixture of learned ones is natural for humans, aiding them to grasp new concepts and strengthening old ones. For all their power and recent success, deep convolutional networks do not have this ability. Inspired by recent work on universal representations for neural...
Hlavní autoři: | Data, G, Ngu, K, Murray, D, Prisacariu, V |
---|---|
Médium: | Conference item |
Vydáno: |
Springer
2018
|
Podobné jednotky
-
Early Earthquake Detection Using Batch Normalization Graph Convolutional Neural Network (BNGCNN)
Autor: Muhammad Atif Bilal, a další
Vydáno: (2022-07-01) -
Convolution neural network with batch normalization and inception-residual modules for Android malware classification
Autor: TianYue Liu, a další
Vydáno: (2022-08-01) -
Development of Deep Convolutional Neural Network with Adaptive Batch Normalization Algorithm for Bearing Fault Diagnosis
Autor: Chao Fu, a další
Vydáno: (2020-01-01) -
Multiple Sclerosis Identification by 14-Layer Convolutional Neural Network With Batch Normalization, Dropout, and Stochastic Pooling
Autor: Shui-Hua Wang, a další
Vydáno: (2018-11-01) -
Reconstruction and Localization of Tumors in Breast Optical Imaging via Convolution Neural Network Based on Batch Normalization Layers
Autor: Nazish Murad, a další
Vydáno: (2022-01-01)