Deep neural network with FGL for small dataset classification
Chunsheng Guo, Ruizhe Li, Meng Yang, Xianghong Tang · IET Image Processing · 2018
In certain applications, classification models have to be trained with small datasets. This study proposes a new deep neural network with a feature generalisation layer (FGL). First, instead of using a generative network for data augmentation, the FGL is modelled using a latent variable model to diversify features directly by sharing other layers. Then, dual‐objective functions are defined to optimise the parameters of the network: one minimises the generation error and the other minimises the classification error. Finally, a parallel multibranch structure is used in the FGL to improve the convergence of model training. The classification accuracy obtained using various quantities of training samples increased up to 4.63% on the MNIST dataset, up to 3.00% on the CIFAR10 nature image dataset, over the reference model. These experimental results illustrate the effectiveness of the authors’ method for training classification models with small datasets.