On the Generalization Error Bounds of Neural Networks under Diversity-Inducing Mutual Angular Regularization
arXiv:1511.07110
Abstract
Recently diversity-inducing regularization methods for latent variable models (LVMs), which encourage the components in LVMs to be diverse, have been studied to address several issues involved in latent variable modeling: (1) how to capture long-tail patterns underlying data; (2) how to reduce model complexity without sacrificing expressivity; (3) how to improve the interpretability of learned patterns. While the effectiveness of diversity-inducing regularizers such as the mutual angular regularizer has been demonstrated empirically, a rigorous theoretical analysis of them is still missing. In this paper, we aim to bridge this gap and analyze how the mutual angular regularizer (MAR) affects the generalization performance of supervised LVMs. We use neural network (NN) as a model instance to carry out the study and the analysis shows that increasing the diversity of hidden units in NN would reduce estimation error and increase approximation error. In addition to theoretical analysis, we also present empirical study which demonstrates that the MAR can greatly improve the performance of NN and the empirical observations are in accordance with the theoretical analysis.
References in corpus (2)
Cited by in corpus (11)
- Diversity in Machine Learning
- Generalization in Deep Learning
- Simple Question Answering by Attentive Convolutional Neural Network
- Understanding Generalization through Visualizations
- Exclusivity Regularized Machine
- Why and How to Pay Different Attention to Phrase Alignments of Different Intensities
- Two Dimensional Stochastic Configuration Networks for Image Data Analytics
- Deep Semi-Random Features for Nonlinear Function Approximation
- Faster Convergence & Generalization in DNNs
- Radius-margin bounds for deep neural networks
- Diversity Regularized Adversarial Learning