pith. the verified trust layer for science. sign in

arxiv: 1907.06257 · v1 · pith:QVHR54JZnew · submitted 2019-07-14 · 💻 cs.LG · cs.CC· stat.ML

More Supervision, Less Computation: Statistical-Computational Tradeoffs in Weakly Supervised Learning

classification 💻 cs.LG cs.CCstat.ML
keywords alphacomputationalalgorithmssupervisionaccuracystatisticalachievingboundaries
0
0 comments X p. Extension
Add this Pith Number to your LaTeX paper What is a Pith Number?
\usepackage{pith}
\pithnumber{QVHR54JZ}

Prints a linked pith:QVHR54JZ badge after your title and writes the identifier into PDF metadata. Compiles on arXiv with no extra files. Learn more

read the original abstract

We consider the weakly supervised binary classification problem where the labels are randomly flipped with probability $1- {\alpha}$. Although there exist numerous algorithms for this problem, it remains theoretically unexplored how the statistical accuracies and computational efficiency of these algorithms depend on the degree of supervision, which is quantified by ${\alpha}$. In this paper, we characterize the effect of ${\alpha}$ by establishing the information-theoretic and computational boundaries, namely, the minimax-optimal statistical accuracy that can be achieved by all algorithms, and polynomial-time algorithms under an oracle computational model. For small ${\alpha}$, our result shows a gap between these two boundaries, which represents the computational price of achieving the information-theoretic boundary due to the lack of supervision. Interestingly, we also show that this gap narrows as ${\alpha}$ increases. In other words, having more supervision, i.e., more correct labels, not only improves the optimal statistical accuracy as expected, but also enhances the computational efficiency for achieving such accuracy.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.