1 paper
Takumi Kawashima, Qing Yu, Akari Asai +2
We propose a new optimization framework for aleatoric uncertainty estimation in regression problems. Existing methods can quantify the error in the target estimation, but they tend…