Kernel Recursive ABC: Point Estimation with Intractable Likelihood

Takafumi Kajihara, Motonobu Kanagawa, Keisuke Yamazaki, Kenji Fukumizu
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:2400-2409, 2018.

Abstract

We propose a novel approach to parameter estimation for simulator-based statistical models with intractable likelihood. Our proposed method involves recursive application of kernel ABC and kernel herding to the same observed data. We provide a theoretical explanation regarding why the approach works, showing (for the population setting) that, under a certain assumption, point estimates obtained with this method converge to the true parameter, as recursion proceeds. We have conducted a variety of numerical experiments, including parameter estimation for a real-world pedestrian flow simulator, and show that in most cases our method outperforms existing approaches.

Cite this Paper

BibTeX
@InProceedings{pmlr-v80-kajihara18a, title = {Kernel Recursive {ABC}: Point Estimation with Intractable Likelihood}, author = {Kajihara, Takafumi and Kanagawa, Motonobu and Yamazaki, Keisuke and Fukumizu, Kenji}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {2400--2409}, 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/kajihara18a/kajihara18a.pdf}, url = {https://proceedings.mlr.press/v80/kajihara18a.html}, abstract = {We propose a novel approach to parameter estimation for simulator-based statistical models with intractable likelihood. Our proposed method involves recursive application of kernel ABC and kernel herding to the same observed data. We provide a theoretical explanation regarding why the approach works, showing (for the population setting) that, under a certain assumption, point estimates obtained with this method converge to the true parameter, as recursion proceeds. We have conducted a variety of numerical experiments, including parameter estimation for a real-world pedestrian flow simulator, and show that in most cases our method outperforms existing approaches.} }
Endnote
%0 Conference Paper %T Kernel Recursive ABC: Point Estimation with Intractable Likelihood %A Takafumi Kajihara %A Motonobu Kanagawa %A Keisuke Yamazaki %A Kenji Fukumizu %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-kajihara18a %I PMLR %P 2400--2409 %U https://proceedings.mlr.press/v80/kajihara18a.html %V 80 %X We propose a novel approach to parameter estimation for simulator-based statistical models with intractable likelihood. Our proposed method involves recursive application of kernel ABC and kernel herding to the same observed data. We provide a theoretical explanation regarding why the approach works, showing (for the population setting) that, under a certain assumption, point estimates obtained with this method converge to the true parameter, as recursion proceeds. We have conducted a variety of numerical experiments, including parameter estimation for a real-world pedestrian flow simulator, and show that in most cases our method outperforms existing approaches.
APA
Kajihara, T., Kanagawa, M., Yamazaki, K. & Fukumizu, K.. (2018). Kernel Recursive ABC: Point Estimation with Intractable Likelihood. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:2400-2409 Available from https://proceedings.mlr.press/v80/kajihara18a.html.

Related Material