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

Model-based deep reinforcement learning has achieved success in variousdomains that require high sample efficiencies, such as Go and robotics.However, there are some remaining issues, such as planning efficientexplorations to learn more accurate dynamic models, evaluating the uncertaintyof the learned models, and more rational utilization of models. To mitigatethese issues, we present MEEE, a model-ensemble method that consists ofoptimistic exploration and weighted exploitation. During exploration, unlikeprior methods directly selecting the optimal action that maximizes the expectedaccumulative return, our agent first generates a set of action candidates andthen seeks out the optimal action that takes both expected return and futureobservation novelty into account. During exploitation, different discountedweights are assigned to imagined transition tuples according to their modeluncertainty respectively, which will prevent model predictive error propagationin agent training. Experiments on several challenging continuous controlbenchmark tasks demonstrated that our approach outperforms other model-free andmodel-based state-of-the-art methods, especially in sample complexity.

Sample Efficient Reinforcement Learning via Model-Ensemble Exploration and Exploitation