Regression with Sensor Data Containing Incomplete Observations

Takayuki Katsuki, Takayuki Osogami
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:15911-15927, 2023.

Abstract

This paper addresses a regression problem in which output label values are the results of sensing the magnitude of a phenomenon. A low value of such labels can mean either that the actual magnitude of the phenomenon was low or that the sensor made an incomplete observation. This leads to a bias toward lower values in labels and the resultant learning because labels may have lower values due to incomplete observations, even if the actual magnitude of the phenomenon was high. Moreover, because an incomplete observation does not provide any tags indicating incompleteness, we cannot eliminate or impute them. To address this issue, we propose a learning algorithm that explicitly models incomplete observations corrupted with an asymmetric noise that always has a negative value. We show that our algorithm is unbiased as if it were learned from uncorrupted data that does not involve incomplete observations. We demonstrate the advantages of our algorithm through numerical experiments.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-katsuki23a, title = {Regression with Sensor Data Containing Incomplete Observations}, author = {Katsuki, Takayuki and Osogami, Takayuki}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {15911--15927}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/katsuki23a/katsuki23a.pdf}, url = {https://proceedings.mlr.press/v202/katsuki23a.html}, abstract = {This paper addresses a regression problem in which output label values are the results of sensing the magnitude of a phenomenon. A low value of such labels can mean either that the actual magnitude of the phenomenon was low or that the sensor made an incomplete observation. This leads to a bias toward lower values in labels and the resultant learning because labels may have lower values due to incomplete observations, even if the actual magnitude of the phenomenon was high. Moreover, because an incomplete observation does not provide any tags indicating incompleteness, we cannot eliminate or impute them. To address this issue, we propose a learning algorithm that explicitly models incomplete observations corrupted with an asymmetric noise that always has a negative value. We show that our algorithm is unbiased as if it were learned from uncorrupted data that does not involve incomplete observations. We demonstrate the advantages of our algorithm through numerical experiments.} }
Endnote
%0 Conference Paper %T Regression with Sensor Data Containing Incomplete Observations %A Takayuki Katsuki %A Takayuki Osogami %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-katsuki23a %I PMLR %P 15911--15927 %U https://proceedings.mlr.press/v202/katsuki23a.html %V 202 %X This paper addresses a regression problem in which output label values are the results of sensing the magnitude of a phenomenon. A low value of such labels can mean either that the actual magnitude of the phenomenon was low or that the sensor made an incomplete observation. This leads to a bias toward lower values in labels and the resultant learning because labels may have lower values due to incomplete observations, even if the actual magnitude of the phenomenon was high. Moreover, because an incomplete observation does not provide any tags indicating incompleteness, we cannot eliminate or impute them. To address this issue, we propose a learning algorithm that explicitly models incomplete observations corrupted with an asymmetric noise that always has a negative value. We show that our algorithm is unbiased as if it were learned from uncorrupted data that does not involve incomplete observations. We demonstrate the advantages of our algorithm through numerical experiments.
APA
Katsuki, T. & Osogami, T.. (2023). Regression with Sensor Data Containing Incomplete Observations. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:15911-15927 Available from https://proceedings.mlr.press/v202/katsuki23a.html.

Related Material