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

Model-free deep reinforcement learning algorithms have been shown to becapable of learning a wide range of robotic skills, but typically require avery large number of samples to achieve good performance. Model-basedalgorithms, in principle, can provide for much more efficient learning, buthave proven difficult to extend to expressive, high-capacity models such asdeep neural networks. In this work, we demonstrate that medium-sized neuralnetwork models can in fact be combined with model predictive control (MPC) toachieve excellent sample complexity in a model-based reinforcement learningalgorithm, producing stable and plausible gaits to accomplish various complexlocomotion tasks. We also propose using deep neural network dynamics models toinitialize a model-free learner, in order to combine the sample efficiency ofmodel-based approaches with the high task-specific performance of model-freemethods. We empirically demonstrate on MuJoCo locomotion tasks that our puremodel-based approach trained on just random action data can follow arbitrarytrajectories with excellent sample efficiency, and that our hybrid algorithmcan accelerate model-free learning on high-speed benchmark tasks, achievingsample efficiency gains of 3-5x on swimmer, cheetah, hopper, and ant agents.Videos can be found at https://sites.google.com/view/mbmf

Neural Network Dynamics for Model-Based Deep Reinforcement Learning with Model-Free Fine-Tuning