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

Deep reinforcement learning with domain randomization learns a control policyin various simulations with randomized physical and sensor model parameters tobecome transferable to the real world in a zero-shot setting. However, a hugenumber of samples are often required to learn an effective policy when therange of randomized parameters is extensive due to the instability of policyupdates. To alleviate this problem, we propose a sample-efficient method namedcyclic policy distillation (CPD). CPD divides the range of randomizedparameters into several small sub-domains and assigns a local policy to eachone. Then local policies are learned while cyclically transitioning tosub-domains. CPD accelerates learning through knowledge transfer based onexpected performance improvements. Finally, all of the learned local policiesare distilled into a global policy for sim-to-real transfers. CPD'seffectiveness and sample efficiency are demonstrated through simulations withfour tasks (Pendulum from OpenAIGym and Pusher, Swimmer, and HalfCheetah fromMujoco), and a real-robot, ball-dispersal task. We published code and videosfrom our experiments athttps://github.com/yuki-kadokawa/cyclic-policy-distillation.

Cyclic Policy Distillation: Sample-Efficient Sim-to-Real Reinforcement Learning with Domain Randomization