Corpus ID: 3204312

Accelerated Training for Matrix-norm Regularization: A Boosting Approach

@inproceedings{Zhang2012AcceleratedTF,
  title={Accelerated Training for Matrix-norm Regularization: A Boosting Approach},
  author={X. Zhang and Y. Yu and Dale Schuurmans},
  booktitle={NIPS},
  year={2012}
}

X. Zhang, Y. Yu, Dale Schuurmans

Published in NIPS 2012

Computer Science, Mathematics

Sparse learning models typically combine a smooth loss with a nonsmooth penalty, such as trace norm. Although recent developments in sparse approximation have offered promising solution methods, current approaches either apply only to matrix-norm constrained problems or provide suboptimal convergence rates. In this paper, we propose a boosting method for regularized learning that guarantees e accuracy within O(1 /e) iterations. Performance is further accelerated by interlacing boosting with… CONTINUE READING

View Paper

cs.uic.edu

Save to Library

Create Alert

Cite

Launch Research Feed