Large Neural Networks Learning from Scratch with Very Few Data and without Explicit Regularization

TL;DR

Highly over-parameterized neural networks can learn effectively from very few data points without explicit regularization or pretraining, challenging traditional machine learning assumptions.

Contribution

This study empirically demonstrates that large CNNs can learn from limited data without regularization across multiple fine-grained classification datasets.

Findings

Large CNNs achieve high accuracy with minimal data without explicit regularization.

Over-parameterized models memorize training data yet generalize well in practice.

Randomly initialized VGG19 can distinguish classes with high accuracy using few samples.

Abstract

Recent findings have shown that highly over-parameterized Neural Networks generalize without pretraining or explicit regularization. It is achieved with zero training error, i.e., complete over-fitting by memorizing the training data. This is surprising, since it is completely against traditional machine learning wisdom. In our empirical study we fortify these findings in the domain of fine-grained image classification. We show that very large Convolutional Neural Networks with millions of weights do learn with only a handful of training samples and without image augmentation, explicit regularization or pretraining. We train the architectures ResNet018, ResNet101 and VGG19 on subsets of the difficult benchmark datasets Caltech101, CUB_200_2011, FGVCAircraft, Flowers102 and StanfordCars with 100 classes and more, perform a comprehensive comparative study and draw implications for the practical application of CNNs. Finally, we show that a randomly initialized VGG19 with 140 million weights learns to distinguish airplanes and motorbikes with up to 95% accuracy using only 20 training samples per class.