Large Deviations for Accelerating Neural Networks Training

Sreelekha Guggilam, Varun Chandola, Abani Patra

Artificial neural networks (ANNs) require tremendous amount of data to train on. However, in classification models, most data features are often similar which can lead to increase…