On improving deep learning generalization with adaptive sparse connectivity
This addresses generalization issues in deep learning for practitioners dealing with limited data, though it is incremental as it builds on existing sparse training methods.
The paper tackles the problem of limited generalization in deep neural networks with limited data by showing that intrinsically sparse networks with adaptive connectivity have better generalization than fully-connected networks, achieving competitive classification performance on 15 datasets while zeroing out around 50% of hidden neurons during training.
Large neural networks are very successful in various tasks. However, with limited data, the generalization capabilities of deep neural networks are also very limited. In this paper, we empirically start showing that intrinsically sparse neural networks with adaptive sparse connectivity, which by design have a strict parameter budget during the training phase, have better generalization capabilities than their fully-connected counterparts. Besides this, we propose a new technique to train these sparse models by combining the Sparse Evolutionary Training (SET) procedure with neurons pruning. Operated on MultiLayer Perceptron (MLP) and tested on 15 datasets, our proposed technique zeros out around 50% of the hidden neurons during training, while having a linear number of parameters to optimize with respect to the number of neurons. The results show a competitive classification and generalization performance.