Co-occurring keywords · papers mentioning both
Papers
A Universal Growth Rate for Learning with Smooth Surrogate Losses
NeurIPS 2024
Is Score Matching Suitable for Estimating Point Processes?
NeurIPS 2024
Realizable H-Consistent and Bayes-Consistent Loss Functions for Learning to Defer
NeurIPS 2024
Regression with Multi-Expert Deferral
ICML 2024
H-Consistency Bounds: Characterization and Extensions
NeurIPS 2023
Cross-Entropy Loss Functions: Theoretical Analysis and Applications
ICML 2023
Structured Prediction with Stronger Consistency Guarantees
NeurIPS 2023
H-Consistency Bounds for Surrogate Loss Minimizers
ICML 2022