Structure Identification by Optimized Interventions

Alberto Giovanni Busetto, Joachim Buhmann
Proceedings of the Twelfth International Conference on Artificial Intelligence and Statistics, PMLR 5:57-64, 2009.

Abstract

We consider the problem of optimal experimental design in structure identification. Whereas standard approaches simply minimize Shannon’s entropy of the estimated parameter posterior, we show how to select between alternative model configurations, too. Our method specifies the intervention that makes an experiment capable of determining whether or not a particular configuration hypothesis is correct. This is performed by a novel clustering technique in approximated Bayesian parameter estimation for non-linear dynamical systems. The computation of the perturbation that minimizes the effective number of clusters in the belief state is constrained by the increase of the expected Kullback-Leibler divergence between the parameter prior and the posterior. This enables the disambiguation of persisting alternative explanations in cases where standard design systematically fails. Its applicability is illustrated with a biochemical Goodwin model, showing correct identification between multiple kinetic structures. We expect that our approach will prove useful especially for complex structures with reduced observability and multimodal posteriors.

Cite this Paper


BibTeX
@InProceedings{pmlr-v5-busetto09a, title = {Structure Identification by Optimized Interventions}, author = {Busetto, Alberto Giovanni and Buhmann, Joachim}, booktitle = {Proceedings of the Twelfth International Conference on Artificial Intelligence and Statistics}, pages = {57--64}, year = {2009}, editor = {van Dyk, David and Welling, Max}, volume = {5}, series = {Proceedings of Machine Learning Research}, address = {Hilton Clearwater Beach Resort, Clearwater Beach, Florida USA}, month = {16--18 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v5/busetto09a/busetto09a.pdf}, url = {https://proceedings.mlr.press/v5/busetto09a.html}, abstract = {We consider the problem of optimal experimental design in structure identification. Whereas standard approaches simply minimize Shannon’s entropy of the estimated parameter posterior, we show how to select between alternative model configurations, too. Our method specifies the intervention that makes an experiment capable of determining whether or not a particular configuration hypothesis is correct. This is performed by a novel clustering technique in approximated Bayesian parameter estimation for non-linear dynamical systems. The computation of the perturbation that minimizes the effective number of clusters in the belief state is constrained by the increase of the expected Kullback-Leibler divergence between the parameter prior and the posterior. This enables the disambiguation of persisting alternative explanations in cases where standard design systematically fails. Its applicability is illustrated with a biochemical Goodwin model, showing correct identification between multiple kinetic structures. We expect that our approach will prove useful especially for complex structures with reduced observability and multimodal posteriors.} }
Endnote
%0 Conference Paper %T Structure Identification by Optimized Interventions %A Alberto Giovanni Busetto %A Joachim Buhmann %B Proceedings of the Twelfth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2009 %E David van Dyk %E Max Welling %F pmlr-v5-busetto09a %I PMLR %P 57--64 %U https://proceedings.mlr.press/v5/busetto09a.html %V 5 %X We consider the problem of optimal experimental design in structure identification. Whereas standard approaches simply minimize Shannon’s entropy of the estimated parameter posterior, we show how to select between alternative model configurations, too. Our method specifies the intervention that makes an experiment capable of determining whether or not a particular configuration hypothesis is correct. This is performed by a novel clustering technique in approximated Bayesian parameter estimation for non-linear dynamical systems. The computation of the perturbation that minimizes the effective number of clusters in the belief state is constrained by the increase of the expected Kullback-Leibler divergence between the parameter prior and the posterior. This enables the disambiguation of persisting alternative explanations in cases where standard design systematically fails. Its applicability is illustrated with a biochemical Goodwin model, showing correct identification between multiple kinetic structures. We expect that our approach will prove useful especially for complex structures with reduced observability and multimodal posteriors.
RIS
TY - CPAPER TI - Structure Identification by Optimized Interventions AU - Alberto Giovanni Busetto AU - Joachim Buhmann BT - Proceedings of the Twelfth International Conference on Artificial Intelligence and Statistics DA - 2009/04/15 ED - David van Dyk ED - Max Welling ID - pmlr-v5-busetto09a PB - PMLR DP - Proceedings of Machine Learning Research VL - 5 SP - 57 EP - 64 L1 - http://proceedings.mlr.press/v5/busetto09a/busetto09a.pdf UR - https://proceedings.mlr.press/v5/busetto09a.html AB - We consider the problem of optimal experimental design in structure identification. Whereas standard approaches simply minimize Shannon’s entropy of the estimated parameter posterior, we show how to select between alternative model configurations, too. Our method specifies the intervention that makes an experiment capable of determining whether or not a particular configuration hypothesis is correct. This is performed by a novel clustering technique in approximated Bayesian parameter estimation for non-linear dynamical systems. The computation of the perturbation that minimizes the effective number of clusters in the belief state is constrained by the increase of the expected Kullback-Leibler divergence between the parameter prior and the posterior. This enables the disambiguation of persisting alternative explanations in cases where standard design systematically fails. Its applicability is illustrated with a biochemical Goodwin model, showing correct identification between multiple kinetic structures. We expect that our approach will prove useful especially for complex structures with reduced observability and multimodal posteriors. ER -
APA
Busetto, A.G. & Buhmann, J.. (2009). Structure Identification by Optimized Interventions. Proceedings of the Twelfth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 5:57-64 Available from https://proceedings.mlr.press/v5/busetto09a.html.

Related Material