Rectify Heterogeneous Models with Semantic Mapping

Han-Jia Ye, De-Chuan Zhan, Yuan Jiang, Zhi-Hua Zhou
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:5630-5639, 2018.

Abstract

On the way to the robust learner for real-world applications, there are still great challenges, including considering unknown environments with limited data. Learnware (Zhou; 2016) describes a novel perspective, and claims that learning models should have reusable and evolvable properties. We propose to Encode Meta InformaTion of features (EMIT), as the model specification for characterizing the changes, which grants the model evolvability to bridge heterogeneous feature spaces. Then, pre-trained models from related tasks can be Reused by our REctiFy via heterOgeneous pRedictor Mapping (REFORM}) framework. In summary, the pre-trained model is adapted to a new environment with different features, through model refining on only a small amount of training data in the current task. Experimental results over both synthetic and real-world tasks with diverse feature configurations validate the effectiveness and practical utility of the proposed framework.

Cite this Paper


BibTeX
@InProceedings{pmlr-v80-ye18c, title = {Rectify Heterogeneous Models with Semantic Mapping}, author = {Ye, Han-Jia and Zhan, De-Chuan and Jiang, Yuan and Zhou, Zhi-Hua}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {5630--5639}, 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/ye2018c/ye2018c.pdf}, url = {https://proceedings.mlr.press/v80/ye18c.html}, abstract = {On the way to the robust learner for real-world applications, there are still great challenges, including considering unknown environments with limited data. Learnware (Zhou; 2016) describes a novel perspective, and claims that learning models should have reusable and evolvable properties. We propose to Encode Meta InformaTion of features (EMIT), as the model specification for characterizing the changes, which grants the model evolvability to bridge heterogeneous feature spaces. Then, pre-trained models from related tasks can be Reused by our REctiFy via heterOgeneous pRedictor Mapping (REFORM}) framework. In summary, the pre-trained model is adapted to a new environment with different features, through model refining on only a small amount of training data in the current task. Experimental results over both synthetic and real-world tasks with diverse feature configurations validate the effectiveness and practical utility of the proposed framework.} }
Endnote
%0 Conference Paper %T Rectify Heterogeneous Models with Semantic Mapping %A Han-Jia Ye %A De-Chuan Zhan %A Yuan Jiang %A Zhi-Hua Zhou %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-ye18c %I PMLR %P 5630--5639 %U https://proceedings.mlr.press/v80/ye18c.html %V 80 %X On the way to the robust learner for real-world applications, there are still great challenges, including considering unknown environments with limited data. Learnware (Zhou; 2016) describes a novel perspective, and claims that learning models should have reusable and evolvable properties. We propose to Encode Meta InformaTion of features (EMIT), as the model specification for characterizing the changes, which grants the model evolvability to bridge heterogeneous feature spaces. Then, pre-trained models from related tasks can be Reused by our REctiFy via heterOgeneous pRedictor Mapping (REFORM}) framework. In summary, the pre-trained model is adapted to a new environment with different features, through model refining on only a small amount of training data in the current task. Experimental results over both synthetic and real-world tasks with diverse feature configurations validate the effectiveness and practical utility of the proposed framework.
APA
Ye, H., Zhan, D., Jiang, Y. & Zhou, Z.. (2018). Rectify Heterogeneous Models with Semantic Mapping. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:5630-5639 Available from https://proceedings.mlr.press/v80/ye18c.html.

Related Material