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

System identification aims to build models of dynamical systems from data.Traditionally, choosing the model requires the designer to balance between twogoals of conflicting nature; the model must be rich enough to capture thesystem dynamics, but not so flexible that it learns spurious random effectsfrom the dataset. It is typically observed that the model validationperformance follows a U-shaped curve as the model complexity increases. Recentdevelopments in machine learning and statistics, however, have observedsituations where a "double-descent" curve subsumes this U-shapedmodel-performance curve. With a second decrease in performance occurring beyondthe point where the model has reached the capacity of interpolating - i.e.,(near) perfectly fitting - the training data. To the best of our knowledge,such phenomena have not been studied within the context of dynamic systems. Thepresent paper aims to answer the question: "Can such a phenomenon also beobserved when estimating parameters of dynamic systems?" We show that theanswer is yes, verifying such behavior experimentally both for artificiallygenerated and real-world datasets.

Beyond Occam's Razor in System Identification: Double-Descent when Modeling Dynamics