From Pixels to Perception: Interpretable Predictions via Instance-wise Grouped Feature Selection

Moritz Vandenhirtz, Julia E Vogt
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:60833-60856, 2025.

Abstract

Understanding the decision-making process of machine learning models provides valuable insights into the task, the data, and the reasons behind a model’s failures. In this work, we propose a method that performs inherently interpretable predictions through the instance-wise sparsification of input images. To align the sparsification with human perception, we learn the masking in the space of semantically meaningful pixel regions rather than on pixel-level. Additionally, we introduce an explicit way to dynamically determine the required level of sparsity for each instance. We show empirically on semi-synthetic and natural image datasets that our inherently interpretable classifier produces more meaningful, human-understandable predictions than state-of-the-art benchmarks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-vandenhirtz25a, title = {From Pixels to Perception: Interpretable Predictions via Instance-wise Grouped Feature Selection}, author = {Vandenhirtz, Moritz and Vogt, Julia E}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {60833--60856}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/vandenhirtz25a/vandenhirtz25a.pdf}, url = {https://proceedings.mlr.press/v267/vandenhirtz25a.html}, abstract = {Understanding the decision-making process of machine learning models provides valuable insights into the task, the data, and the reasons behind a model’s failures. In this work, we propose a method that performs inherently interpretable predictions through the instance-wise sparsification of input images. To align the sparsification with human perception, we learn the masking in the space of semantically meaningful pixel regions rather than on pixel-level. Additionally, we introduce an explicit way to dynamically determine the required level of sparsity for each instance. We show empirically on semi-synthetic and natural image datasets that our inherently interpretable classifier produces more meaningful, human-understandable predictions than state-of-the-art benchmarks.} }
Endnote
%0 Conference Paper %T From Pixels to Perception: Interpretable Predictions via Instance-wise Grouped Feature Selection %A Moritz Vandenhirtz %A Julia E Vogt %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-vandenhirtz25a %I PMLR %P 60833--60856 %U https://proceedings.mlr.press/v267/vandenhirtz25a.html %V 267 %X Understanding the decision-making process of machine learning models provides valuable insights into the task, the data, and the reasons behind a model’s failures. In this work, we propose a method that performs inherently interpretable predictions through the instance-wise sparsification of input images. To align the sparsification with human perception, we learn the masking in the space of semantically meaningful pixel regions rather than on pixel-level. Additionally, we introduce an explicit way to dynamically determine the required level of sparsity for each instance. We show empirically on semi-synthetic and natural image datasets that our inherently interpretable classifier produces more meaningful, human-understandable predictions than state-of-the-art benchmarks.
APA
Vandenhirtz, M. & Vogt, J.E.. (2025). From Pixels to Perception: Interpretable Predictions via Instance-wise Grouped Feature Selection. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:60833-60856 Available from https://proceedings.mlr.press/v267/vandenhirtz25a.html.

Related Material