Neural Networks Regularization Through Class-wise Invariant Representation Learning
This addresses the challenge of data scarcity in neural network training for classification tasks, though it is incremental as it builds on existing regularization methods.
The paper tackles the problem of training neural networks for classification with limited training samples by introducing a regularization term that enforces class-wise invariant representations in hidden layers, and shows improved generalization on MNIST variants, particularly with few samples.
Training deep neural networks is known to require a large number of training samples. However, in many applications only few training samples are available. In this work, we tackle the issue of training neural networks for classification task when few training samples are available. We attempt to solve this issue by proposing a new regularization term that constrains the hidden layers of a network to learn class-wise invariant representations. In our regularization framework, learning invariant representations is generalized to the class membership where samples with the same class should have the same representation. Numerical experiments over MNIST and its variants showed that our proposal helps improving the generalization of neural network particularly when trained with few samples. We provide the source code of our framework https://github.com/sbelharbi/learning-class-invariant-features .