Deep Learning with a Rethinking Structure for Multi-label Classification

Yao-Yuan Yang, Yi-An Lin, Hong-Min Chu, Hsuan-Tien Lin
Proceedings of The Eleventh Asian Conference on Machine Learning, PMLR 101:125-140, 2019.

Abstract

Multi-label classification (MLC) is an important class of machine learning problems that come with a wide spectrum of applications, each demanding a possibly different evaluation criterion. When solving the MLC problems, we generally expect the learning algorithm to take the hidden correlation of the labels into account to improve the prediction performance. Extracting the hidden correlation is generally a challenging task. In this work, we propose a novel deep learning framework to better extract the hidden correlation with the help of the memory structure within recurrent neural networks. The memory stores the temporary guesses on the labels and effectively allows the framework to rethink about the goodness and correlation of the guesses before making the final prediction. Furthermore, the rethinking process makes it easy to adapt to different evaluation criteria to match real-world application needs. In particular, the framework can be trained in an end-to-end style with respect to any given MLC evaluation criteria. The end-to-end design can be seamlessly combined with other deep learning techniques to conquer challenging MLC problems like image tagging. Experimental results across many real-world data sets justify that the rethinking framework indeed improves MLC performance across different evaluation criteria and leads to superior performance over state-of-the-art MLC algorithms.

Cite this Paper

BibTeX
@InProceedings{pmlr-v101-yang19b, title = {Deep Learning with a Rethinking Structure for Multi-label Classification}, author = {Yang, Yao-Yuan and Lin, Yi-An and Chu, Hong-Min and Lin, Hsuan-Tien}, booktitle = {Proceedings of The Eleventh Asian Conference on Machine Learning}, pages = {125--140}, year = {2019}, editor = {Lee, Wee Sun and Suzuki, Taiji}, volume = {101}, series = {Proceedings of Machine Learning Research}, month = {17--19 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v101/yang19b/yang19b.pdf}, url = {https://proceedings.mlr.press/v101/yang19b.html}, abstract = {Multi-label classification (MLC) is an important class of machine learning problems that come with a wide spectrum of applications, each demanding a possibly different evaluation criterion. When solving the MLC problems, we generally expect the learning algorithm to take the hidden correlation of the labels into account to improve the prediction performance. Extracting the hidden correlation is generally a challenging task. In this work, we propose a novel deep learning framework to better extract the hidden correlation with the help of the memory structure within recurrent neural networks. The memory stores the temporary guesses on the labels and effectively allows the framework to rethink about the goodness and correlation of the guesses before making the final prediction. Furthermore, the rethinking process makes it easy to adapt to different evaluation criteria to match real-world application needs. In particular, the framework can be trained in an end-to-end style with respect to any given MLC evaluation criteria. The end-to-end design can be seamlessly combined with other deep learning techniques to conquer challenging MLC problems like image tagging. Experimental results across many real-world data sets justify that the rethinking framework indeed improves MLC performance across different evaluation criteria and leads to superior performance over state-of-the-art MLC algorithms.} }
Endnote
%0 Conference Paper %T Deep Learning with a Rethinking Structure for Multi-label Classification %A Yao-Yuan Yang %A Yi-An Lin %A Hong-Min Chu %A Hsuan-Tien Lin %B Proceedings of The Eleventh Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2019 %E Wee Sun Lee %E Taiji Suzuki %F pmlr-v101-yang19b %I PMLR %P 125--140 %U https://proceedings.mlr.press/v101/yang19b.html %V 101 %X Multi-label classification (MLC) is an important class of machine learning problems that come with a wide spectrum of applications, each demanding a possibly different evaluation criterion. When solving the MLC problems, we generally expect the learning algorithm to take the hidden correlation of the labels into account to improve the prediction performance. Extracting the hidden correlation is generally a challenging task. In this work, we propose a novel deep learning framework to better extract the hidden correlation with the help of the memory structure within recurrent neural networks. The memory stores the temporary guesses on the labels and effectively allows the framework to rethink about the goodness and correlation of the guesses before making the final prediction. Furthermore, the rethinking process makes it easy to adapt to different evaluation criteria to match real-world application needs. In particular, the framework can be trained in an end-to-end style with respect to any given MLC evaluation criteria. The end-to-end design can be seamlessly combined with other deep learning techniques to conquer challenging MLC problems like image tagging. Experimental results across many real-world data sets justify that the rethinking framework indeed improves MLC performance across different evaluation criteria and leads to superior performance over state-of-the-art MLC algorithms.
APA
Yang, Y., Lin, Y., Chu, H. & Lin, H.. (2019). Deep Learning with a Rethinking Structure for Multi-label Classification. Proceedings of The Eleventh Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 101:125-140 Available from https://proceedings.mlr.press/v101/yang19b.html.

Related Material