Adaptive Dropout Rates for Learning with Corrupted Features
Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence. Special Track on Machine Learning. Pages 4126-4132.
Abstract
Feature noising is an effective mechanism on reducing the risk of overfitting. To avoid an explosive searching space, existing work typically assumes that all features share a single noise level, which is often cross-validated. In this paper, we present a Bayesian feature noising model that flexibly allows for dimension-specific or group-specific noise levels, and we derive a learning algorithm that adaptively updates these noise levels. Our adaptive rule is simple and interpretable, by drawing a direct connection to the fitness of each individual feature or feature group. Empirical results on various datasets demonstrate the effectiveness on avoiding extensive tuning and sometimes improving the performance due to its flexibility.
@inproceedings{ijcai2015p579,
title = {Adaptive Dropout Rates for Learning with Corrupted Features},
author = {Zhuo, Jingwei and Zhu, Jun and Zhang, Bo},
booktitle = {Proceedings of the Twenty-Fourth International Joint Conference on
Artificial Intelligence, {IJCAI-15}},
publisher = {International Joint Conferences on Artificial Intelligence},
editor = {Qiang Yang and Michael Wooldridge},
pages = {4126--4132},
year = {2015},
month = {7},
note = {Special Track on Machine Learning},
url = {https://ijcai.org/proceedings/2015/579},
}
TY - CPAPER TI - Adaptive Dropout Rates for Learning with Corrupted Features AU - Zhuo, Jingwei AU - Zhu, Jun AU - Zhang, Bo T2 - Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-15 PB - International Joint Conferences on Artificial Intelligence PY - 2015 SP - 4126 EP - 4132 UR - https://ijcai.org/proceedings/2015/579 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
