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

Learning from demonstration is an effective method for human users toinstruct desired robot behaviour. However, for most non-trivial tasks ofpractical interest, efficient learning from demonstration depends crucially oninductive bias in the chosen structure for rewards/costs and policies. Weaddress the case where this inductive bias comes from an exchange with a humanuser. We propose a method in which a learning agent utilizes the informationbottleneck layer of a high-parameter variational neural model, with auxiliaryloss terms, in order to ground abstract concepts such as spatial relations. Theconcepts are referred to in natural language instructions and are manifested inthe high-dimensional sensory input stream the agent receives from the world. Weevaluate the properties of the latent space of the learned model in aphotorealistic synthetic environment and particularly focus on examining itsusability for downstream tasks. Additionally, through a series of controlledtable-top manipulation experiments, we demonstrate that the learned manifoldcan be used to ground demonstrations as symbolic plans, which can then beexecuted on a PR2 robot.

Disentangled Relational Representations for Explaining and Learning from Demonstration