Adapting Multicomponent Predictive Systems using Hybrid Adaptation Strategies with Auto-WEKA in Process Industry

Manuel Martin Salvador, Marcin Budka, Bogdan Gabrys
Proceedings of the Workshop on Automatic Machine Learning, PMLR 64:48-57, 2016.

Abstract

Automation of composition and optimisation of multicomponent predictive systems (MCPSs) made of a number of preprocessing steps and predictive models is a challenging problem that has been addressed in recent works. However, one of the current challenges is how to adapt these systems in dynamic environments where data is changing over time. In this work we propose a hybrid approach combining different adaptation strategies with the Bayesian optimisation techniques for parametric, structural and hyperparameter optimisation of entire MCPSs. Experiments comparing different adaptation strategies have been performed on 7 datasets from real chemical production processes. Experimental analysis shows that optimisation of entire MCPSs as a method of adaptation to changing environments is feasible and that hybrid strategies perform better in most of the analysed cases.

Cite this Paper

BibTeX
@InProceedings{pmlr-v64-salvador_adapting_2016, title = {Adapting Multicomponent Predictive Systems using Hybrid Adaptation Strategies with Auto-WEKA in Process Industry}, author = {Salvador, Manuel Martin and Budka, Marcin and Gabrys, Bogdan}, booktitle = {Proceedings of the Workshop on Automatic Machine Learning}, pages = {48--57}, year = {2016}, editor = {Hutter, Frank and Kotthoff, Lars and Vanschoren, Joaquin}, volume = {64}, series = {Proceedings of Machine Learning Research}, address = {New York, New York, USA}, month = {24 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v64/salvador_adapting_2016.pdf}, url = {https://proceedings.mlr.press/v64/salvador_adapting_2016.html}, abstract = {Automation of composition and optimisation of multicomponent predictive systems (MCPSs) made of a number of preprocessing steps and predictive models is a challenging problem that has been addressed in recent works. However, one of the current challenges is how to adapt these systems in dynamic environments where data is changing over time. In this work we propose a hybrid approach combining different adaptation strategies with the Bayesian optimisation techniques for parametric, structural and hyperparameter optimisation of entire MCPSs. Experiments comparing different adaptation strategies have been performed on 7 datasets from real chemical production processes. Experimental analysis shows that optimisation of entire MCPSs as a method of adaptation to changing environments is feasible and that hybrid strategies perform better in most of the analysed cases.} }
Endnote
%0 Conference Paper %T Adapting Multicomponent Predictive Systems using Hybrid Adaptation Strategies with Auto-WEKA in Process Industry %A Manuel Martin Salvador %A Marcin Budka %A Bogdan Gabrys %B Proceedings of the Workshop on Automatic Machine Learning %C Proceedings of Machine Learning Research %D 2016 %E Frank Hutter %E Lars Kotthoff %E Joaquin Vanschoren %F pmlr-v64-salvador_adapting_2016 %I PMLR %P 48--57 %U https://proceedings.mlr.press/v64/salvador_adapting_2016.html %V 64 %X Automation of composition and optimisation of multicomponent predictive systems (MCPSs) made of a number of preprocessing steps and predictive models is a challenging problem that has been addressed in recent works. However, one of the current challenges is how to adapt these systems in dynamic environments where data is changing over time. In this work we propose a hybrid approach combining different adaptation strategies with the Bayesian optimisation techniques for parametric, structural and hyperparameter optimisation of entire MCPSs. Experiments comparing different adaptation strategies have been performed on 7 datasets from real chemical production processes. Experimental analysis shows that optimisation of entire MCPSs as a method of adaptation to changing environments is feasible and that hybrid strategies perform better in most of the analysed cases.
RIS
TY - CPAPER TI - Adapting Multicomponent Predictive Systems using Hybrid Adaptation Strategies with Auto-WEKA in Process Industry AU - Manuel Martin Salvador AU - Marcin Budka AU - Bogdan Gabrys BT - Proceedings of the Workshop on Automatic Machine Learning DA - 2016/12/04 ED - Frank Hutter ED - Lars Kotthoff ED - Joaquin Vanschoren ID - pmlr-v64-salvador_adapting_2016 PB - PMLR DP - Proceedings of Machine Learning Research VL - 64 SP - 48 EP - 57 L1 - http://proceedings.mlr.press/v64/salvador_adapting_2016.pdf UR - https://proceedings.mlr.press/v64/salvador_adapting_2016.html AB - Automation of composition and optimisation of multicomponent predictive systems (MCPSs) made of a number of preprocessing steps and predictive models is a challenging problem that has been addressed in recent works. However, one of the current challenges is how to adapt these systems in dynamic environments where data is changing over time. In this work we propose a hybrid approach combining different adaptation strategies with the Bayesian optimisation techniques for parametric, structural and hyperparameter optimisation of entire MCPSs. Experiments comparing different adaptation strategies have been performed on 7 datasets from real chemical production processes. Experimental analysis shows that optimisation of entire MCPSs as a method of adaptation to changing environments is feasible and that hybrid strategies perform better in most of the analysed cases. ER -
APA
Salvador, M.M., Budka, M. & Gabrys, B.. (2016). Adapting Multicomponent Predictive Systems using Hybrid Adaptation Strategies with Auto-WEKA in Process Industry. Proceedings of the Workshop on Automatic Machine Learning, in Proceedings of Machine Learning Research 64:48-57 Available from https://proceedings.mlr.press/v64/salvador_adapting_2016.html.

Related Material