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/

Zendy Plus

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

Zendy Tools

Zendy Open

We present a method for fast training of vision based control policies onreal robots. The key idea behind our method is to perform multi-taskReinforcement Learning with auxiliary tasks that differ not only in the rewardto be optimized but also in the state-space in which they operate. Inparticular, we allow auxiliary task policies to utilize task features that areavailable only at training-time. This allows for fast learning of auxiliarypolicies, which subsequently generate good data for training the main,vision-based control policies. This method can be seen as an extension of theScheduled Auxiliary Control (SAC-X) framework. We demonstrate the efficacy ofour method by using both a simulated and real-world Ball-in-a-Cup gamecontrolled by a robot arm. In simulation, our approach leads to significantlearning speed-ups when compared to standard SAC-X. On the real robot we showthat the task can be learned from-scratch, i.e., with no transfer fromsimulation and no imitation learning. Videos of our learned policies running onthe real robot can be found athttps://sites.google.com/view/rss-2019-sawyer-bic/.

Simultaneously Learning Vision and Feature-based Control Policies for Real-world Ball-in-a-Cup