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 adoption of Reinforcement Learning (RL) in several human-centredapplications provides robots with autonomous decision-making capabilities andadaptability based on the observations of the operating environment. In suchscenarios, however, the learning process can make robots' behaviours unclearand unpredictable to humans, thus preventing a smooth and effective Human-RobotInteraction (HRI). As a consequence, it becomes crucial to avoid robotsperforming actions that are unclear to the user. In this work, we investigatewhether including human preferences in RL (concerning the actions the robotperforms during learning) improves the transparency of a robot's behaviours.For this purpose, a shielding mechanism is included in the RL algorithm toinclude human preferences and to monitor the learning agent's decisions. Wecarried out a within-subjects study involving 26 participants to evaluate therobot's transparency in terms of Legibility, Predictability, and Expectabilityin different settings. Results indicate that considering human preferencesduring learning improves Legibility with respect to providing onlyExplanations, and combining human preferences with explanations elucidating therationale behind the robot's decisions further amplifies transparency. Resultsalso confirm that an increase in transparency leads to an increase in thesafety, comfort, and reliability of the robot. These findings show theimportance of transparency during learning and suggest a paradigm for roboticapplications with human in the loop.

Increasing Transparency of Reinforcement Learning using Shielding for Human Preferences and Explanations