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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Creating open-ended algorithms, which generate their own never-ending streamof novel and appropriately challenging learning opportunities, could help toautomate and accelerate progress in machine learning. A recent step in thisdirection is the Paired Open-Ended Trailblazer (POET), an algorithm thatgenerates and solves its own challenges, and allows solutions to goal-switchbetween challenges to avoid local optima. However, the original POET was unableto demonstrate its full creative potential because of limitations of thealgorithm itself and because of external issues including a limited problemspace and lack of a universal progress measure. Importantly, both limitationspose impediments not only for POET, but for the pursuit of open-endedness ingeneral. Here we introduce and empirically validate two new innovations to theoriginal algorithm, as well as two external innovations designed to helpelucidate its full potential. Together, these four advances enable the mostopen-ended algorithmic demonstration to date. The algorithmic innovations are(1) a domain-general measure of how meaningfully novel new challenges are,enabling the system to potentially create and solve interesting challengesendlessly, and (2) an efficient heuristic for determining when agents shouldgoal-switch from one problem to another (helping open-ended search betterscale). Outside the algorithm itself, to enable a more definitive demonstrationof open-endedness, we introduce (3) a novel, more flexible way to encodeenvironmental challenges, and (4) a generic measure of the extent to which asystem continues to exhibit open-ended innovation. Enhanced POET produces adiverse range of sophisticated behaviors that solve a wide range ofenvironmental challenges, many of which cannot be solved through other means.

Enhanced POET: Open-Ended Reinforcement Learning through Unbounded Invention of Learning Challenges and their Solutions