A multi-task loss mixing rating classification, regression, and learned uncertainty weights improves reward learning from ratings over a classification-only baseline in synthetic control tasks.
Multi-task learning using uncertainty to weigh losses for scene geometry and semantics
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Multi-Task Reward Learning from Human Ratings
A multi-task loss mixing rating classification, regression, and learned uncertainty weights improves reward learning from ratings over a classification-only baseline in synthetic control tasks.