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

Self-paced reinforcement learning (RL) aims to improve the data efficiency oflearning by automatically creating sequences, namely curricula, of probabilitydistributions over contexts. However, existing techniques for self-paced RLfail in long-horizon planning tasks that involve temporally extended behaviors.We hypothesize that taking advantage of prior knowledge about the underlyingtask structure can improve the effectiveness of self-paced RL. We develop aself-paced RL algorithm guided by reward machines, i.e., a type of finite-statemachine that encodes the underlying task structure. The algorithm integratesreward machines in 1) the update of the policy and value functions obtained byany RL algorithm of choice, and 2) the update of the automated curriculum thatgenerates context distributions. Our empirical results evidence that theproposed algorithm achieves optimal behavior reliably even in cases in whichexisting baselines cannot make any meaningful progress. It also decreases thecurriculum length and reduces the variance in the curriculum generation processby up to one-fourth and four orders of magnitude, respectively.

Reward-Machine-Guided, Self-Paced Reinforcement Learning