Mastering Model Fit: Applications of the InterModel Vigorish for Interpretable Machine Learning
Abstract
Accuracy alone gives little information about which features or model components improve probabilistic fit, which outcomes benefit, or how fit changes during training. We introduce the InterModel Vigorish (IMV) into the Machine Intelligence space in order to express differences in fit as an expected betting advantage over a reference model. SHAP-IMV attributes gains to features through Shapley-weighted subset comparisons, Multi-IMV evaluates one-vs-rest and pairwise class discrimination, and Ablate-IMV assesses architectural components and feature groups by ablation. We evaluate these applications over ten random seeds using tabular data, images, sensor sequences, and graphs. Feature rankings vary across estimators. High one-vs-rest accuracy can accompany negligible improvement over a null, while ablation can preserve or improve fit under a fixed training budget. On six graph-classification benchmarks, we time-profile IMV over 200 training epochs, reporting seed means and ranges. The open-source imvpy library and accompanying notebooks provide code to replicate our work and to use the IMV more widely.
Related
- Variable-Projection Sparse Functional Principal Component Analysis: Interpretable Functional Dimensionality Reduction with Applications to Raman Spectral Data
- Multiscale Nonlinear Risk Transmission in the Carbon–Fossil Energy–Clean Energy System from a Fractal-Market Perspective: Evidence from EMD–TVP-VAR and Interpretable Machine Learning
- HARMONIA: Interpretable Graph Learning through Mixtures of Neural Bases
- Null importance: Disentangling relevance for interpretable machine learning
- Predicting mental health risk in adolescents using interpretable machine learning: Evidence from SHAP analysis
BibTeX
@misc{yan2026mastering,
title = {{Mastering Model Fit: Applications of the InterModel Vigorish for Interpretable Machine Learning}},
author = {Jiani Yan and Pengyun Wang and Klint Kanopka and Ben Domingue and Charles Rahal},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.29960},
eprint = {2610.29960},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.29960},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - Mastering Model Fit: Applications of the InterModel Vigorish for Interpretable Machine Learning
AU - Jiani Yan
AU - Pengyun Wang
AU - Klint Kanopka
AU - Ben Domingue
AU - Charles Rahal
PY - 2026
DA - 2026/10/21
PB - arXiv
DO - 10.48550/arxiv.2610.29960
UR - https://arxiv.org/abs/2610.29960
AB - Accuracy alone gives little information about which features or model components improve probabilistic fit, which outcomes benefit, or how fit changes during training. We introduce the InterModel Vigorish (IMV) into the Machine Intelligence space in order to express differences in fit as an expected betting advantage over a reference model. SHAP-IMV attributes gains to features through Shapley-weighted subset comparisons, Multi-IMV evaluates one-vs-rest and pairwise class discrimination, and Ablate-IMV assesses architectural components and feature groups by ablation. We evaluate these applications over ten random seeds using tabular data, images, sensor sequences, and graphs. Feature rankings vary across estimators. High one-vs-rest accuracy can accompany negligible improvement over a null, while ablation can preserve or improve fit under a fixed training budget. On six graph-classification benchmarks, we time-profile IMV over 200 training epochs, reporting seed means and ranges. The open-source imvpy library and accompanying notebooks provide code to replicate our work and to use the IMV more widely.
ER -