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

Probabilistic regression techniques in control and robotics applications haveto fulfill different criteria of data-driven adaptability, computationalefficiency, scalability to high dimensions, and the capacity to deal withdifferent modalities in the data. Classical regressors usually fulfill only asubset of these properties. In this work, we extend seminal work on Bayesiannonparametric mixtures and derive an efficient variational Bayes inferencetechnique for infinite mixtures of probabilistic local polynomial models withwell-calibrated certainty quantification. We highlight the model's power incombining data-driven complexity adaptation, fast prediction and the ability todeal with discontinuous functions and heteroscedastic noise. We benchmark thistechnique on a range of large real inverse dynamics datasets, showing that theinfinite mixture formulation is competitive with classical Local Learningmethods and regularizes model complexity by adapting the number of componentsbased on data and without relying on heuristics. Moreover, to showcase thepracticality of the approach, we use the learned models for online inversedynamics control of a Barrett-WAM manipulator, significantly improving thetrajectory tracking performance.

A Variational Infinite Mixture for Probabilistic Inverse Dynamics Learning