2019/09/21 by Beyrem Khalfaoui, Khalfaoui, Beyrem, J. T. Boyd +3
Computer Science · #FOS: Computer and information sciences #Gaussian Processes and Bayesian Inference #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning and ELM #Neural Networks and Applications
paper · pdf · doi:10.48550/arxiv.1909.09819
openalex publication_date 2019/09/21 · openalex created_date 2022/07/28 · openalex updated_date 2026/07/28
Dropout is a regularisation technique in neural network training where unit activations are randomly set to zero with a given probability independently. In this work, we propose a generalisation of dropout and other multiplicative noise injection schemes for shallow and deep neural networks, where the random noise applied to different units is not independent but follows a joint distribution that is either fixed or estimated during training. We provide theoretical insights on why such adaptive structured noise injection (ASNI) may be relevant, and empirically confirm that it helps boost the accuracy of simple feedforward and convolutional neural networks, disentangles the hidden layer representations, and leads to sparser representations. Our proposed method is a straightforward modification of the classical dropout and does not require additional computational overhead.