10 papers
Coarsening Latent-Class Probabilities: Directional Distortion and Coverage Loss
Marcell T. Kurbucz
Outcomes are increasingly regressed on a calibrated probability vector for unobserved class membership, and that vector is often coarsened to a hard label first. Under a constant-c…
Adaptive Conditional Forest Sampling for Spectral Risk Optimisation under Decision-Dependent Uncertainty
Marcell T. Kurbucz
Minimising a spectral risk objective, defined as a weighted combination of expected cost and Conditional Value-at-Risk (CVaR), is challenging when the uncertainty distribution is d…
When to Trust Confidence Thresholding: Calibration Diagnostics for Pseudo-Labelled Regression
Marcell T. Kurbucz
Calibrated probability outputs of trained classifiers are increasingly used as inputs to downstream regression estimands such as effects, prevalences, or disparities for a latent g…
Identification of Latent Group Effects under Conditional Calibration
Marcell T. Kurbucz
We study identification of a structural group effect when the group indicator is unobserved, but the analyst observes a calibrated probability score satisfying $E…
SplitWise Regression: Stepwise Modeling with Adaptive Dummy Encoding
Marcell T. Kurbucz, Nikolaos Tzivanakis, Nilufer Sari Aslam +1
Capturing nonlinear relationships without sacrificing interpretability remains a persistent challenge in regression modeling. We introduce SplitWise, a novel framework that enhance…
ALT: A Python Package for Lightweight Feature Representation in Time Series Classification
Balázs P. Halmos, Balázs Hajós, Vince Ã. Molnár +2
We introduce ALT, an open-source Python package created for efficient and accurate time series classification (TSC). The package implements the adaptive law-based transformation (A…