Self-paced Mixture of Regression

We introduce self-paced learning into mixture of regressions, proposing a novel self-paced regularizer based on Exclusive LASSO that alleviates intra-component outliers and inter-component data imbalance, achieving superior performance on age estimation and glucose estimation.

摘要

Mixture of regressions (MoR) is the well-established and effective approach to model discontinuous and heterogeneous data in regression problems. Existing MoR approaches assume smooth joint distribution for its good analytic properties. However, such assumption makes existing MoR very sensitive to intra-component outliers (the noisy training data residing in certain components) and the inter-component imbalance (the different amounts of training data in different components). In this paper, we make the earliest effort on Self-paced Learning (SPL) in MoR, i.e., Self-paced mixture of regressions (SPMoR) model. We propose a novel self-paced regularizer based on the Exclusive LASSO, which improves inter-component balance of training data. As a robust learning regime, SPL pursues confidence sample reasoning. To demonstrate the effectiveness of SPMoR, we conducted experiments on both the synthetic examples and real-world applications to age estimation and glucose estimation.

主要结果

BibTeX

@inproceedings{han2017spmor,
  author    = {Han, Longfei and Zhang, Dingwen and Huang, Dong and Chang, Xiaojun and Han, Junwei},
  title     = {Self-paced Mixture of Regression},
  booktitle = {Proceedings of the 26th International Joint Conference on Artificial Intelligence (IJCAI)},
  year      = {2017},
  doi       = {10.24963/ijcai.2017/252}
}