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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Interactions with either environments or expert policies during training areneeded for most of the current imitation learning (IL) algorithms. For ILproblems with no interactions, a typical approach is Behavior Cloning (BC).However, BC-like methods tend to be affected by distribution shift. To mitigatethis problem, we come up with a Robust Model-Based Imitation Learning (RMBIL)framework that casts imitation learning as an end-to-end differentiablenonlinear closed-loop tracking problem. RMBIL applies Neural ODE to learn aprecise multi-step dynamics and a robust tracking controller via NonlinearDynamics Inversion (NDI) algorithm. Then, the learned NDI controller will becombined with a trajectory generator, a conditional VAE, to imitate an expert'sbehavior. Theoretical derivation shows that the controller network canapproximate an NDI when minimizing the training loss of Neural ODE. Experimentson Mujoco tasks also demonstrate that RMBIL is competitive to thestate-of-the-art generative adversarial method (GAIL) and achieves at least 30%performance gain over BC in uneven surfaces.

No Need for Interactions: Robust Model-Based Imitation Learning using Neural ODE