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

Rather than learning new control policies for each new task, it is possible,when tasks share some structure, to compose a "meta-policy" from previouslylearned policies. This paper reports results from experiments using DeepReinforcement Learning on a continuous-state, discrete-action autonomousdriving simulator. We explore how Deep Neural Networks can representmeta-policies that switch among a set of previously learned policies,specifically in settings where the dynamics of a new scenario are composed of amixture of previously learned dynamics and where the state observation ispossibly corrupted by sensing noise. We also report the results of experimentsvarying dynamics mixes, distractor policies, magnitudes/distributions ofsensing noise, and obstacles. In a fully observed experiment, the meta-policylearning algorithm achieves 2.6x the reward achieved by the next best policycomposition technique with 80% less exploration. In a partially observedexperiment, the meta-policy learning algorithm converges after 50 iterationswhile a direct application of RL fails to converge even after 200 iterations.

Composing Meta-Policies for Autonomous Driving Using Hierarchical Deep Reinforcement Learning