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 this paper we study a new reinforcement learning setting where theenvironment is non-rewarding, contains several possibly related objects ofvarious controllability, and where an apt agent Bob acts independently, withnon-observable intentions. We argue that this setting defines a realisticscenario and we present a generic discrete-state discrete-action model of suchenvironments. To learn in this environment, we propose an unsupervisedreinforcement learning agent called CLIC for Curriculum Learning and Imitationfor Control. CLIC learns to control individual objects in its environment, andimitates Bob's interactions with these objects. It selects objects to focus onwhen training and imitating by maximizing its learning progress. We show thatCLIC is an effective baseline in our new setting. It can effectively observeBob to gain control of objects faster, even if Bob is not explicitly teaching.It can also follow Bob when he acts as a mentor and provides ordereddemonstrations. Finally, when Bob controls objects that the agent cannot, or inpresence of a hierarchy between objects in the environment, we show that CLICignores non-reproducible and already mastered interactions with objects,resulting in a greater benefit from imitation.

CLIC: Curriculum Learning and Imitation for object Control in non-rewarding environments