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

Recent advances have shown the capability of Fully Convolutional NeuralNetworks (FCN) to model cost functions for motion planning in the context oflearning driving preferences purely based on demonstration data from humandrivers. While pure learning from demonstrations in the framework of InverseReinforcement Learning (IRL) is a promising approach, we can benefit from wellinformed human priors and incorporate them into the learning process. Our workachieves this by pretraining a model to regress to a manual cost function andrefining it based on Maximum Entropy Deep Inverse Reinforcement Learning. Wheninjecting prior knowledge as pretraining for the network, we achieve higherrobustness, more visually distinct obstacle boundaries, and the ability tocapture instances of obstacles that elude models that purely learn fromdemonstration data. Furthermore, by exploiting these human priors, theresulting model can more accurately handle corner cases that are scarcely seenin the demonstration data, such as stairs, slopes, and underpasses.

Incorporating Human Domain Knowledge into Large Scale Cost Function Learning