NeurIPS 2016poster19 citations

Fast learning rates with heavy-tailed losses

Vu C Dinh, Lam S Ho, Binh Nguyen, Duy Nguyen

Abstract

We study fast learning rates when the losses are not necessarily bounded and may have a distribution with heavy tails. To enable such analyses, we introduce two new conditions: (i) the envelope function $\sup_{f \in \mathcal{F}}|\ell \circ f|$, where $\ell$ is the loss function and $\mathcal{F}$ is the hypothesis class, exists and is $L^r$-integrable, and (ii) $\ell$ satisfies the multi-scale Bernstein's condition on $\mathcal{F}$. Under these assumptions, we prove that learning rate faster than $O(n^{-1/2})$ can be obtained and, depending on $r$ and the multi-scale Bernstein's powers, can be arbitrarily close to $O(n^{-1})$. We then verify these assumptions and derive fast learning rates for the problem of vector quantization by $k$-means clustering with heavy-tailed distributions. The analyses enable us to obtain novel learning rates that extend and complement existing results in the literature from both theoretical and practical viewpoints.

BibTeX
@inproceedings{NIPS2016_63923f49,
 author = {Dinh, Vu C and Ho, Lam S and Nguyen, Binh and Nguyen, Duy},
 booktitle = {Advances in Neural Information Processing Systems},
 editor = {D. Lee and M. Sugiyama and U. Luxburg and I. Guyon and R. Garnett},
 pages = {},
 publisher = {Curran Associates, Inc.},
 title = {Fast learning rates with heavy-tailed losses},
 url = {https://proceedings.neurips.cc/paper_files/paper/2016/file/63923f49e5241343aa7acb6a06a751e7-Paper.pdf},
 volume = {29},
 year = {2016}
}
Fast learning rates with heavy-tailed losses · NeurIPS 2016