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

The goal of this paper is to make a strong point for the usage of dynamicalmodels when using reinforcement learning (RL) for feedback control of dynamicalsystems governed by partial differential equations (PDEs). To breach the gapbetween the immense promises we see in RL and the applicability in complexengineering systems, the main challenges are the massive requirements in termsof the training data, as well as the lack of performance guarantees. We presenta solution for the first issue using a data-driven surrogate model in the formof a convolutional LSTM with actuation. We demonstrate that learning anactuated model in parallel to training the RL agent significantly reduces thetotal amount of required data sampled from the real system. Furthermore, weshow that iteratively updating the model is of major importance to avoid biasesin the RL training. Detailed ablation studies reveal the most importantingredients of the modeling process. We use the chaotic Kuramoto-Sivashinskyequation do demonstarte our findings.

Learning a model is paramount for sample efficiency in reinforcement learning control of PDEs