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

Go-Explore achieved breakthrough performance on challenging reinforcementlearning (RL) tasks with sparse rewards. The key insight of Go-Explore was thatsuccessful exploration requires an agent to first return to an interestingstate ('Go'), and only then explore into unknown terrain ('Explore'). We referto such exploration after a goal is reached as 'post-exploration'. In thispaper we present a systematic study of post-exploration, answering openquestions that the Go-Explore paper did not answer yet. First, we study theisolated potential of post-exploration, by turning it on and off within thesame algorithm. Subsequently, we introduce new methodology to adaptively decidewhen to post-explore and for how long to post-explore. Experiments on a rangeof MiniGrid environments show that post-exploration indeed boosts performance(with a bigger impact than tuning regular exploration parameters), and thiseffect is further enhanced by adaptively deciding when and for how long topost-explore. In short, our work identifies adaptive post-exploration as apromising direction for RL exploration research.

When to Go, and When to Explore: The Benefit of Post-Exploration in Intrinsic Motivation