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

In robotics, it is essential to be able to plan efficiently inhigh-dimensional continuous state-action spaces for long horizons. For suchcomplex planning problems, unguided uniform sampling of actions until a path toa goal is found is hopelessly inefficient, and gradient-based approaches oftenfall short when the optimization manifold of a given problem is not smooth. Inthis paper we present an approach that guides the search of a state-spaceplanner, such as A*, by learning an action-sampling distribution that cangeneralize across different instances of a planning problem. The motivation isthat, unlike typical learning approaches for planning for continuous actionspace that estimate a policy, an estimated action sampler is more robust toerror since it has a planner to fall back on. We use a Generative AdversarialNetwork (GAN), and address an important issue: search experience consists of arelatively large number of actions that are not on a solution path and arelatively small number of actions that actually are on a solution path. Weintroduce a new technique, based on an importance-ratio estimation method, forusing samples from a non-target distribution to make GAN learning moredata-efficient. We provide theoretical guarantees and empirical evaluation inthree challenging continuous robot planning problems to illustrate theeffectiveness of our algorithm.

Guiding the search in continuous state-action spaces by learning an action sampling distribution from off-target samples