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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

A challenging problem when studying a dynamical system is to find theinterdependencies among its individual components. Several algorithms have beenproposed to detect directed dynamical influences between time series. Two ofthe most used approaches are a model-free one (transfer entropy) and amodel-based one (Granger causality). Several pitfalls are related to thepresence or absence of assumptions in modeling the relevant features of thedata. We tried to overcome those pitfalls using a neural network approach inwhich a model is built without any a priori assumptions. In this sense thismethod can be seen as a bridge between model-free and model-based approaches.The experiments performed will show that the method presented in this work candetect the correct dynamical information flows occurring in a system of timeseries. Additionally we adopt a non-uniform embedding framework according towhich only the past states that actually help the prediction are entered intothe model, improving the prediction and avoiding the risk of overfitting. Thismethod also leads to a further improvement with respect to traditional Grangercausality approaches when redundant variables (i.e. variables sharing the sameinformation about the future of the system) are involved. Neural networks arealso able to recognize dynamics in data sets completely different from the onesused during the training phase.

Neural Networks with Non-Uniform Embedding and Explicit Validation Phase to Assess Granger Causality