We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Understanding the Disharmony between Weight Normalization Family and Weight Decay: $ε-$shifted $L_2$ Regularizer

Abstract: The merits of fast convergence and potentially better performance of the weight normalization family have drawn increasing attention in recent years. These methods use standardization or normalization that changes the weight $\boldsymbol{W}$ to $\boldsymbol{W}'$, which makes $\boldsymbol{W}'$ independent to the magnitude of $\boldsymbol{W}$. Surprisingly, $\boldsymbol{W}$ must be decayed during gradient descent, otherwise we will observe a severe under-fitting problem, which is very counter-intuitive since weight decay is widely known to prevent deep networks from over-fitting. In this paper, we \emph{theoretically} prove that the weight decay term $\frac{1}{2}\lambda||{\boldsymbol{W}}||^2$ merely modulates the effective learning rate for improving objective optimization, and has no influence on generalization when the weight normalization family is compositely employed. Furthermore, we also expose several critical problems when introducing weight decay term to weight normalization family, including the missing of global minimum and training instability. To address these problems, we propose an $\epsilon-$shifted $L_2$ regularizer, which shifts the $L_2$ objective by a positive constant $\epsilon$. Such a simple operation can theoretically guarantee the existence of global minimum, while preventing the network weights from being too small and thus avoiding gradient float overflow. It significantly improves the training stability and can achieve slightly better performance in our practice. The effectiveness of $\epsilon-$shifted $L_2$ regularizer is comprehensively validated on the ImageNet, CIFAR-100, and COCO datasets. Our codes and pretrained models will be released in this https URL
Comments: 12 pages, 9 figures
Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
Cite as: arXiv:1911.05920 [cs.LG]
  (or arXiv:1911.05920v1 [cs.LG] for this version)

Submission history

From: Xiang Li [view email]
[v1] Thu, 14 Nov 2019 03:31:13 GMT (1104kb,D)

Link back to: arXiv, form interface, contact.