1 paper
Daniel Sikar, Artur Garcez, Robin Bloomfield +6
This study introduces the Misclassification Likelihood Matrix (MLM) as a novel tool for quantifying the reliability of neural network predictions under distribution shifts. The MLM…