Temporal distribution shifts in pharmaceutical assay data, strongest in target-based assays, degrade the calibration of popular uncertainty quantification methods, and post hoc calibration fails when the calibration-test shift is large.
Sources of uncertainty in machine learning – a statisticians’ view, 2023
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.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Temporal Distribution Shift in Real-World Pharmaceutical Data: Implications for Uncertainty Quantification in QSAR Models
Temporal distribution shifts in pharmaceutical assay data, strongest in target-based assays, degrade the calibration of popular uncertainty quantification methods, and post hoc calibration fails when the calibration-test shift is large.