Collaborative Novelty Detection for Distributed Data by a Probabilistic Method

Akira Imakura, Xiucai Ye, Tetsuya Sakurai
Proceedings of The 13th Asian Conference on Machine Learning, PMLR 157:932-947, 2021.

Abstract

Novelty detection, which detects anomalies based on a training dataset consisting of only the normal data, is an important task in several applications. In addition, in the real world, there may be situations where data is owned by multiple parties in a distributed manner but cannot be shared with each other due to privacy and confidentiality requirements. Therefore, how to develop distributed novelty detection while preserving privacy is essential. To address this challenge, we propose a probabilistic collaborative method that allows distributed novelty detection for multiple parties without sharing the original data. The proposed method constructs a collaborative kernel based on a collaborative data analysis framework, by which intermediate representations are generated from each party and shared for collaborative novelty detection. Numerical experiments demonstrate that the proposed method obtains better performance compared with the individual novelty detection in the local party.

Cite this Paper


BibTeX
@InProceedings{pmlr-v157-imakura21a, title = {Collaborative Novelty Detection for Distributed Data by a Probabilistic Method}, author = {Imakura, Akira and Ye, Xiucai and Sakurai, Tetsuya}, booktitle = {Proceedings of The 13th Asian Conference on Machine Learning}, pages = {932--947}, year = {2021}, editor = {Balasubramanian, Vineeth N. and Tsang, Ivor}, volume = {157}, series = {Proceedings of Machine Learning Research}, month = {17--19 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v157/imakura21a/imakura21a.pdf}, url = {https://proceedings.mlr.press/v157/imakura21a.html}, abstract = {Novelty detection, which detects anomalies based on a training dataset consisting of only the normal data, is an important task in several applications. In addition, in the real world, there may be situations where data is owned by multiple parties in a distributed manner but cannot be shared with each other due to privacy and confidentiality requirements. Therefore, how to develop distributed novelty detection while preserving privacy is essential. To address this challenge, we propose a probabilistic collaborative method that allows distributed novelty detection for multiple parties without sharing the original data. The proposed method constructs a collaborative kernel based on a collaborative data analysis framework, by which intermediate representations are generated from each party and shared for collaborative novelty detection. Numerical experiments demonstrate that the proposed method obtains better performance compared with the individual novelty detection in the local party.} }
Endnote
%0 Conference Paper %T Collaborative Novelty Detection for Distributed Data by a Probabilistic Method %A Akira Imakura %A Xiucai Ye %A Tetsuya Sakurai %B Proceedings of The 13th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Vineeth N. Balasubramanian %E Ivor Tsang %F pmlr-v157-imakura21a %I PMLR %P 932--947 %U https://proceedings.mlr.press/v157/imakura21a.html %V 157 %X Novelty detection, which detects anomalies based on a training dataset consisting of only the normal data, is an important task in several applications. In addition, in the real world, there may be situations where data is owned by multiple parties in a distributed manner but cannot be shared with each other due to privacy and confidentiality requirements. Therefore, how to develop distributed novelty detection while preserving privacy is essential. To address this challenge, we propose a probabilistic collaborative method that allows distributed novelty detection for multiple parties without sharing the original data. The proposed method constructs a collaborative kernel based on a collaborative data analysis framework, by which intermediate representations are generated from each party and shared for collaborative novelty detection. Numerical experiments demonstrate that the proposed method obtains better performance compared with the individual novelty detection in the local party.
APA
Imakura, A., Ye, X. & Sakurai, T.. (2021). Collaborative Novelty Detection for Distributed Data by a Probabilistic Method. Proceedings of The 13th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 157:932-947 Available from https://proceedings.mlr.press/v157/imakura21a.html.

Related Material