Hoeffding Concept Bottleneck Models with Applications to Overhead Images
Abstract
Explainability of deep learning algorithms is critical for computer-vision applications with high-stakes decisions. Concept bottleneck models (CBM) have recently shown promising performance to provide explainable and accurate predictions for classification problems, based on a bottleneck of high-level concepts. Existing CBM methods rely on a linear aggregation of the concept scores to compute predictions. However, a large number of concepts is often used in this linear approach, which undermines explainability and favors information leakage. In general, the underlying relation between concepts and output logits is not linear. Therefore, we introduce Hoeffding Concept Bottleneck Models (HCBM), which build on the Hoeffding functional decomposition of gradient-boosted trees to provide non-linear and sparse aggregations of concept scores, and generate compact predictions using prime implicants. HCBM have strong theoretical properties of robustness to interconcept leakage, and outperform sparse linear CBM in practice, as shown in extensive experiments. Beyond classification, HCBM can be adapted to object detection, and we focus on a challenging case with overhead images to show the high performance of HCBM in these settings.
Related
- When Are Concept Bottleneck Model Explanations Faithful and Compact?
- A Sharper Hoeffding Bound for Weighted Sums of Exchangeable Random Variables
- Explaining Heterogeneous Treatment Effects with Causal Forests and the Hoeffding Functional Decomposition
- A short operator proof of Hoeffding inequalities for Markov chains
- Constant-Overhead Injection into Quantum Codes
BibTeX
@misc{benard2026hoeffding,
title = {{Hoeffding Concept Bottleneck Models with Applications to Overhead Images}},
author = {Clément Bénard and Manon Arfib and Christophe Labreuche and Victor Quétu},
howpublished = {HAL},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2606.00082},
eprint = {2606.00082},
archivePrefix = {arXiv},
url = {https://hal.science/hal-05626982v2},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Hoeffding Concept Bottleneck Models with Applications to Overhead Images
AU - Clément Bénard
AU - Manon Arfib
AU - Christophe Labreuche
AU - Victor Quétu
PY - 2026
DA - 2026/10/09
PB - HAL
DO - 10.48550/arxiv.2606.00082
UR - https://hal.science/hal-05626982v2
AB - Explainability of deep learning algorithms is critical for computer-vision applications with high-stakes decisions. Concept bottleneck models (CBM) have recently shown promising performance to provide explainable and accurate predictions for classification problems, based on a bottleneck of high-level concepts. Existing CBM methods rely on a linear aggregation of the concept scores to compute predictions. However, a large number of concepts is often used in this linear approach, which undermines explainability and favors information leakage. In general, the underlying relation between concepts and output logits is not linear. Therefore, we introduce Hoeffding Concept Bottleneck Models (HCBM), which build on the Hoeffding functional decomposition of gradient-boosted trees to provide non-linear and sparse aggregations of concept scores, and generate compact predictions using prime implicants. HCBM have strong theoretical properties of robustness to interconcept leakage, and outperform sparse linear CBM in practice, as shown in extensive experiments. Beyond classification, HCBM can be adapted to object detection, and we focus on a challenging case with overhead images to show the high performance of HCBM in these settings.
ER -