Semi-Supervised Learning on Data Streams via Temporal Label Propagation

Tal Wagner, Sudipto Guha, Shiva Kasiviswanathan, Nina Mishra
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:5095-5104, 2018.

Abstract

We consider the problem of labeling points on a fast-moving data stream when only a small number of labeled examples are available. In our setting, incoming points must be processed efficiently and the stream is too large to store in its entirety. We present a semi-supervised learning algorithm for this task. The algorithm maintains a small synopsis of the stream which can be quickly updated as new points arrive, and labels every incoming point by provably learning from the full history of the stream. Experiments on real datasets validate that the algorithm can quickly and accurately classify points on a stream with a small quantity of labeled examples.

Cite this Paper


BibTeX
@InProceedings{pmlr-v80-wagner18a, title = {Semi-Supervised Learning on Data Streams via Temporal Label Propagation}, author = {Wagner, Tal and Guha, Sudipto and Kasiviswanathan, Shiva and Mishra, Nina}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {5095--5104}, year = {2018}, editor = {Dy, Jennifer and Krause, Andreas}, volume = {80}, series = {Proceedings of Machine Learning Research}, month = {10--15 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v80/wagner18a/wagner18a.pdf}, url = {https://proceedings.mlr.press/v80/wagner18a.html}, abstract = {We consider the problem of labeling points on a fast-moving data stream when only a small number of labeled examples are available. In our setting, incoming points must be processed efficiently and the stream is too large to store in its entirety. We present a semi-supervised learning algorithm for this task. The algorithm maintains a small synopsis of the stream which can be quickly updated as new points arrive, and labels every incoming point by provably learning from the full history of the stream. Experiments on real datasets validate that the algorithm can quickly and accurately classify points on a stream with a small quantity of labeled examples.} }
Endnote
%0 Conference Paper %T Semi-Supervised Learning on Data Streams via Temporal Label Propagation %A Tal Wagner %A Sudipto Guha %A Shiva Kasiviswanathan %A Nina Mishra %B Proceedings of the 35th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jennifer Dy %E Andreas Krause %F pmlr-v80-wagner18a %I PMLR %P 5095--5104 %U https://proceedings.mlr.press/v80/wagner18a.html %V 80 %X We consider the problem of labeling points on a fast-moving data stream when only a small number of labeled examples are available. In our setting, incoming points must be processed efficiently and the stream is too large to store in its entirety. We present a semi-supervised learning algorithm for this task. The algorithm maintains a small synopsis of the stream which can be quickly updated as new points arrive, and labels every incoming point by provably learning from the full history of the stream. Experiments on real datasets validate that the algorithm can quickly and accurately classify points on a stream with a small quantity of labeled examples.
APA
Wagner, T., Guha, S., Kasiviswanathan, S. & Mishra, N.. (2018). Semi-Supervised Learning on Data Streams via Temporal Label Propagation. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:5095-5104 Available from https://proceedings.mlr.press/v80/wagner18a.html.

Related Material