Fotiadis, Stathi, Lino, Mario, Shunlong, Hu, Garasto, Stef, Cantwell, Chris D and Bharath, Anil A (2023) Disentangled generative models for robust prediction of system dynamics. In: Proceedings of the 40th International Conference on Machine Learning. ICML 2023, 202 . Proceedings of Machine Learning Research (PMLR), Honolulu, Hawaii, USA, pp. 10222-10248. ISSN 2640-3498

Preview

PDF (Published manuscript (online conference proceedings)) 42998_GARASTO_Disentangled_generative_models_for_robust_prediction_of_system_dynamics.pdf - Published Version Download (6MB) | Preview

Abstract

The use of deep neural networks for modelling system dynamics is increasingly popular, but long-term prediction accuracy and out-of-distribution generalization still present challenges. In this study, we address these challenges by considering the parameters of dynamical systems as factors of variation of the data and leverage their ground-truth values to disentangle the representations learned by generative models. Our experimental results in phase-space and observation-space dynamics, demonstrate the effectiveness of latent-space supervision in producing disentangled representations, leading to improved long-term prediction accuracy and out-of-distribution robustness.

Item Type:	Conference Proceedings
Title of Proceedings:	Proceedings of the 40th International Conference on Machine Learning
Uncontrolled Keywords:	generative models; dynamical systems; disentanglement; VAE; forecasting; system dynamics
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Faculty / School / Research Centre / Research Group:	Faculty of Engineering & Science Faculty of Engineering & Science > School of Computing & Mathematical Sciences (CMS)
Related URLs:	Publisher Publisher Publisher
Last Modified:	14 Jul 2023 12:48
URI:	http://gala.gre.ac.uk/id/eprint/42998

Actions (login required)

View Item

Downloads