Forward-Backward Splitting for Time-Varying Graphical Models

Federico Tomasi, Veronica Tozzo, Alessandro Verri, Saverio Salzo
Proceedings of the Ninth International Conference on Probabilistic Graphical Models, PMLR 72:475-486, 2018.

Abstract

Gaussian graphical models have received much attention in the last years, due to their flexibility and expression power. However, the optimisation of such complex models suffer from computational issues both in terms of convergence rates and memory requirements. Here, we present a forward-backward splitting (FBS) procedure for Gaussian graphical modelling of multivariate time-series which relies on recent theoretical studies ensuring convergence under mild assumptions. Our experiments show that a FBS-based implementation achieves, with very fast convergence rates, optimal results with respect to ground truth and standard methods for dynamical network inference. Optimisation algorithms which are usually exploited for network inference suffer from drawbacks when considering large sets of unknowns. Particularly for increasing data sets and model complexity, we argue for the use of fast and theoretically sound optimisation algorithms to be significant to the graphical modelling community.

Cite this Paper


BibTeX
@InProceedings{pmlr-v72-tomasi18a, title = {Forward-Backward Splitting for Time-Varying Graphical Models}, author = {Tomasi, Federico and Tozzo, Veronica and Verri, Alessandro and Salzo, Saverio}, booktitle = {Proceedings of the Ninth International Conference on Probabilistic Graphical Models}, pages = {475--486}, year = {2018}, editor = {Kratochvíl, Václav and Studený, Milan}, volume = {72}, series = {Proceedings of Machine Learning Research}, month = {11--14 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v72/tomasi18a/tomasi18a.pdf}, url = {https://proceedings.mlr.press/v72/tomasi18a.html}, abstract = {Gaussian graphical models have received much attention in the last years, due to their flexibility and expression power. However, the optimisation of such complex models suffer from computational issues both in terms of convergence rates and memory requirements. Here, we present a forward-backward splitting (FBS) procedure for Gaussian graphical modelling of multivariate time-series which relies on recent theoretical studies ensuring convergence under mild assumptions. Our experiments show that a FBS-based implementation achieves, with very fast convergence rates, optimal results with respect to ground truth and standard methods for dynamical network inference. Optimisation algorithms which are usually exploited for network inference suffer from drawbacks when considering large sets of unknowns. Particularly for increasing data sets and model complexity, we argue for the use of fast and theoretically sound optimisation algorithms to be significant to the graphical modelling community.} }
Endnote
%0 Conference Paper %T Forward-Backward Splitting for Time-Varying Graphical Models %A Federico Tomasi %A Veronica Tozzo %A Alessandro Verri %A Saverio Salzo %B Proceedings of the Ninth International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2018 %E Václav Kratochvíl %E Milan Studený %F pmlr-v72-tomasi18a %I PMLR %P 475--486 %U https://proceedings.mlr.press/v72/tomasi18a.html %V 72 %X Gaussian graphical models have received much attention in the last years, due to their flexibility and expression power. However, the optimisation of such complex models suffer from computational issues both in terms of convergence rates and memory requirements. Here, we present a forward-backward splitting (FBS) procedure for Gaussian graphical modelling of multivariate time-series which relies on recent theoretical studies ensuring convergence under mild assumptions. Our experiments show that a FBS-based implementation achieves, with very fast convergence rates, optimal results with respect to ground truth and standard methods for dynamical network inference. Optimisation algorithms which are usually exploited for network inference suffer from drawbacks when considering large sets of unknowns. Particularly for increasing data sets and model complexity, we argue for the use of fast and theoretically sound optimisation algorithms to be significant to the graphical modelling community.
APA
Tomasi, F., Tozzo, V., Verri, A. & Salzo, S.. (2018). Forward-Backward Splitting for Time-Varying Graphical Models. Proceedings of the Ninth International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 72:475-486 Available from https://proceedings.mlr.press/v72/tomasi18a.html.

Related Material