Deep Learning on Small Datasets without Pre-Training using Cosine Loss

By Björn Barz, Joachim Denzler. 2019

157 citations recorded by the source corpus.

Two things seem to be indisputable in the contemporary deep learning discourse: 1. The categorical cross-entropy loss after softmax activation is the method of choice for classification. 2. Training a CNN classifier from scratch on small datasets does not work well.In contrast to this, we show that the cosine loss function provides substantially better performance than crossentropy on datasets with only a handful of samples per class. For example, the accuracy achieved on the CUB- 200-2011 dataset without pre-training is by 30% higher than with the cross-entropy loss. Further experiments on other popular datasets confirm our findings. Moreover, we demonstrate that integrating prior knowledge in the form of class hierarchies is straightforward with the cosine loss and improves classification performance further.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Optimization & Training

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Optimization & Training

Citations in the corpus

Hyperspherical Prototype Networks (2019, unknown): “The first consists of one-hot vectors on the C-dimensional simplex for C classes, as proposed in [2, 7].”
LQF: Linear Quadratic Fine-Tuning (2020, methodology): “The key enablers of LQF are simple and known in the literature, although not frequently used: (i) We replace the cross-entropy loss with the mean-squared error loss, making the optimization problem quadratic [13, 18, 3], (ii) we replace ReLU with Leaky-ReLU [33], and (iii) we perform preconditioning”