English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Empirically observed time series in physics, biology, or medicine, arecommonly generated by some underlying dynamical system (DS) which is the targetof scientific interest. There is an increasing interest to harvest machinelearning methods to reconstruct this latent DS in a data-driven, unsupervisedway. In many areas of science it is common to sample time series observationsfrom many data modalities simultaneously, e.g. electrophysiological andbehavioral time series in a typical neuroscience experiment. However, currentmachine learning tools for reconstructing DSs usually focus on just one datamodality. Here we propose a general framework for multi-modal data integrationfor the purpose of nonlinear DS reconstruction and the analysis of cross-modalrelations. This framework is based on dynamically interpretable recurrentneural networks as general approximators of nonlinear DSs, coupled to sets ofmodality-specific decoder models from the class of generalized linear models.Both an expectation-maximization and a variational inference algorithm formodel training are advanced and compared. We show on nonlinear DS benchmarksthat our algorithms can efficiently compensate for too noisy or missinginformation in one data channel by exploiting other channels, and demonstrateon experimental neuroscience data how the algorithm learns to link differentdata domains to the underlying dynamics.

Reconstructing Nonlinear Dynamical Systems from Multi-Modal Time Series