Unmodified mixtures of experts can provide predictive uncertainty estimates via entropy, mutual information, and expert-variance, and these estimates outperform a two-expert average ensemble on conditional correctness metrics under out-of-distribution data.
Uncertainty prediction and cal- ibration using multi-expert gating mechanism
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Extracting Uncertainty Estimates from Mixtures of Experts for Semantic Segmentation
Unmodified mixtures of experts can provide predictive uncertainty estimates via entropy, mutual information, and expert-variance, and these estimates outperform a two-expert average ensemble on conditional correctness metrics under out-of-distribution data.