Ghosh, Aritra and Manwani, Naresh and Sastry, P S (2017) On the Robustness of Decision Tree Learning Under Label Noise. [Book Chapter]
|
PDF
lec_not_com_sci_10234_685-697_2017.pdf - Published Version Download (215kB) | Preview |
Abstract
In most practical problems of classifier learning, the training data suffers from label noise. Most theoretical results on robustness to label noise involve either estimation of noise rates or non-convex optimization. Further, none of these results are applicable to standard decision tree learning algorithms. This paper presents some theoretical analysis to show that, under some assumptions, many popular decision tree learning algorithms are inherently robust to label noise. 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.
| Item Type: | Book Chapter |
|---|---|
| Publisher: | Springer Verlag |
| Additional Information: | The Copyright of the article belongs to the Authors. |
| Keywords: | Decision trees; Label noise; Robust learning; Classification (of information); Convex optimization; Data mining; Decision trees; Optimization; Classifier learning; Decision tree learning; Decision tree learning algorithm; Extensive simulations; Nonconvex optimization; Practical problems; Robust learning; Sample complexity; Learning algorithms |
| Department/Centre: | Division of Electrical Sciences > Electrical Engineering |
| Date Deposited: | 29 May 2022 06:51 |
| Last Modified: | 31 May 2022 06:04 |
| URI: | https://eprints.iisc.ac.in/id/eprint/72594 |
Actions (login required)
 |
View Item |
