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

AI practitioners typically strive to develop the most accurate systems,making an implicit assumption that the AI system will function autonomously.However, in practice, AI systems often are used to provide advice to people indomains ranging from criminal justice and finance to healthcare. In suchAI-advised decision making, humans and machines form a team, where the human isresponsible for making final decisions. But is the most accurate AI the bestteammate? We argue "No" -- predictable performance may be worth a slightsacrifice in AI accuracy. Instead, we argue that AI systems should be trainedin a human-centered manner, directly optimized for team performance. We studythis proposal for a specific type of human-AI teaming, where the human overseerchooses to either accept the AI recommendation or solve the task themselves. Tooptimize the team performance for this setting we maximize the team's expectedutility, expressed in terms of the quality of the final decision, cost ofverifying, and individual accuracies of people and machines. Our experimentswith linear and non-linear models on real-world, high-stakes datasets show thatthe most accuracy AI may not lead to highest team performance and show thebenefit of modeling teamwork during training through improvements in expectedteam utility across datasets, considering parameters such as human skill andthe cost of mistakes. We discuss the shortcoming of current optimizationapproaches beyond well-studied loss functions such as log-loss, and encouragefuture work on AI optimization problems motivated by human-AI collaboration.

Is the Most Accurate AI the Best Teammate? Optimizing AI for Teamwork