no code implementations • 28 Nov 2023 • Hermanni Hälvä, Jonathan So, Richard E. Turner, Aapo Hyvärinen
In this paper, we introduce a new nonlinear ICA framework that employs $t$-process (TP) latent components which apply naturally to data with higher-dimensional dependency structures, such as spatial and spatio-temporal data.
1 code implementation • NeurIPS 2021 • Hermanni Hälvä, Sylvain Le Corff, Luc Lehéricy, Jonathan So, Yongjie Zhu, Elisabeth Gassiat, Aapo Hyvarinen
We introduce a new general identifiable framework for principled disentanglement referred to as Structured Nonlinear Independent Component Analysis (SNICA).
1 code implementation • 22 Jun 2020 • Hermanni Hälvä, Aapo Hyvärinen
The central idea in such works is that the latent components are assumed to be independent conditional on some observed auxiliary variables, such as the time-segment index.
no code implementations • 19 Jun 2020 • Hiroshi Morioka, Hermanni Hälvä, Aapo Hyvärinen
Additivity greatly limits the generality of the model, hindering analysis of general NVAR processes which have nonlinear interactions between the innovations.
no code implementations • 5 Dec 2018 • Niko Reunanen, Ville Könönen, Hermanni Hälvä, Jani Mäntyjärvi, Arttu Lämsä, Jussi Liikka
The resulting model learns 1) a representation of the atomic activities of a sport and 2) to classify physical activities as compositions of the atomic activities.