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

Humans quickly solve tasks in novel systems with complex dynamics, withoutrequiring much interaction. While deep reinforcement learning algorithms haveachieved tremendous success in many complex tasks, these algorithms need alarge number of samples to learn meaningful policies. In this paper, we presenta task for navigating a marble to the center of a circular maze. While thissystem is very intuitive and easy for humans to solve, it can be very difficultand inefficient for standard reinforcement learning algorithms to learnmeaningful policies. We present a model that learns to move a marble in thecomplex environment within minutes of interacting with the real system.Learning consists of initializing a physics engine with parameters estimatedusing data from the real system. The error in the physics engine is thencorrected using Gaussian process regression, which is used to model theresidual between real observations and physics engine simulations. The physicsengine augmented with the residual model is then used to control the marble inthe maze environment using a model-predictive feedback over a receding horizon.To the best of our knowledge, this is the first time that a hybrid modelconsisting of a full physics engine along with a statistical functionapproximator has been used to control a complex physical system in real-timeusing nonlinear model-predictive control (NMPC).

Data-Efficient Learning for Complex and Real-Time Physical Problem Solving using Augmented Simulation