Auxiliary Data Selection in Percolative Learning Method for Improving Neural Network Performance

Masayuki Kobayashi, Shinichi Shirakawa, Tomoharu Nagao

2022

Abstract

Neural networks have been evolved significantly at the cost of requiring many input data. However, collecting useful data is expensive for many practical uses, which can be barrier for practical use in real-world applications. In this work, we propose a framework for improving the model performance, in which the model leverages the auxiliary data that is only available during the training. We demonstrate how to (i) train the neural network to perform as though auxiliary data are used during the testing, and (ii) automatically select the auxiliary data during training to encourages the model to generalize well and avoid overfitting to the auxiliary data. We evaluate our method on several datasets, and compare the performance with baseline model. Despite the simplicity of our method, our method makes it possible to get good generalization performance in most cases.

Paper Citation

in Harvard Style

Kobayashi M., Shirakawa S. and Nagao T. (2022). Auxiliary Data Selection in Percolative Learning Method for Improving Neural Network Performance. In Proceedings of the 14th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART, ISBN 978-989-758-547-0, pages 381-387. DOI: 10.5220/0010825700003116

in Bibtex Style

@conference{icaart22,
author={Masayuki Kobayashi and Shinichi Shirakawa and Tomoharu Nagao},
title={Auxiliary Data Selection in Percolative Learning Method for Improving Neural Network Performance},
booktitle={Proceedings of the 14th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART,},
year={2022},
pages={381-387},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010825700003116},
isbn={978-989-758-547-0},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART,
TI - Auxiliary Data Selection in Percolative Learning Method for Improving Neural Network Performance
SN - 978-989-758-547-0
AU - Kobayashi M.
AU - Shirakawa S.
AU - Nagao T.
PY - 2022
SP - 381
EP - 387
DO - 10.5220/0010825700003116