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

Artificial intelligence systems increasingly involve continual learning toenable flexibility in general situations that are not encountered during systemtraining. Human interaction with autonomous systems is broadly studied, butresearch has hitherto under-explored interactions that occur while the systemis actively learning, and can noticeably change its behaviour in minutes. Inthis pilot study, we investigate how the interaction between a human and acontinually learning prediction agent develops as the agent developscompetency. Additionally, we compare two different agent architectures toassess how representational choices in agent design affect the human-agentinteraction. We develop a virtual reality environment and a time-basedprediction task wherein learned predictions from a reinforcement learning (RL)algorithm augment human predictions. We assess how a participant's performanceand behaviour in this task differs across agent types, using both quantitativeand qualitative analyses. Our findings suggest that human trust of the systemmay be influenced by early interactions with the agent, and that trust in turnaffects strategic behaviour, but limitations of the pilot study rule out anyconclusive statement. We identify trust as a key feature of interaction tofocus on when considering RL-based technologies, and make severalrecommendations for modification to this study in preparation for alarger-scale investigation. A video summary of this paper can be found athttps://youtu.be/oVYJdnBqTwQ .

Assessing Human Interaction in Virtual Reality With Continually Learning Prediction Agents Based on Reinforcement Learning Algorithms: A Pilot Study