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

Download:

Current browse context:

cs.LG

Change to browse by:

cs

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: On the Robustness of Decision Tree Learning under Label Noise

Abstract: In most practical problems of classifier learning, the training data suffers from the label noise. Hence, it is important to understand how robust is a learning algorithm to such label noise. This paper presents some theoretical analysis to show that many popular decision tree algorithms are robust to symmetric label noise under large sample size. We also present some sample complexity results which provide some bounds on the sample size for the robustness to hold with a high probability. Through extensive simulations we illustrate this robustness.
Subjects: Machine Learning (cs.LG)
Cite as: arXiv:1605.06296 [cs.LG]
  (or arXiv:1605.06296v2 [cs.LG] for this version)

Submission history

From: Naresh Manwani [view email]
[v1] Fri, 20 May 2016 11:31:26 GMT (32kb)
[v2] Fri, 26 Aug 2016 08:58:06 GMT (58kb)

Link back to: arXiv, form interface, contact.