Open access
Feb 2025
What Should an AI Assessor Optimise for?
This work experimentally explores the question: is it always optimal to train the assessor for the target metric, or could it be better to train for a different metric and then map predictions back to the target metric?
Daniel Romero-Alvarado, Fernando Mart'inez-Plumed, José Hernández-Orallo
· Machine-mediated learning · 0 citations