But that's not why: Inference adjustment by interactive prototype revision
Not on the map. Why?
Abstract (arXiv)
Despite significant advances in machine learning, decision-making of artificial agents is still not perfect and often requires post-hoc human interventions. If the prediction of a model relies on unreasonable factors it is desirable to remove their effect. Deep interactive prototype adjustment enables the user to give hints and correct the model's reasoning. In this paper, we demonstrate that prototypical-part models are well suited for this task as their prediction is based on prototypical image patches that can be interpreted semantically by the user. It shows that even correct classifications can rely on unreasonable prototypes that result from confounding variables in a dataset. Hence, we propose simple yet effective interaction schemes for inference adjustment: The user is consulted interactively to identify faulty prototypes. Non-object prototypes can be removed by prototype masking or a custom mode of deselection training. Interactive prototype rejection allows machine learning na"ive users to adjust the logic of reasoning without compromising the accuracy.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 14 of its 25 references that are in the map. First posted on arXiv 18 Mar 2022.
Builds on
14 of 25 references.
References · 14
- 2017Prototypical Networks for Few-shot Learningcited by 3,001
- 2017Deep Bayesian Active Learning with Image Datacited by 812
- 2018This Looks Like That: Deep Learning for Interpretable Image Recognitioncited by 731
- 2020Understanding the Role of Individual Units in a Deep Neural Networkcited by 254
- 2017Deep Learning for Case-Based Reasoning through Prototypes: A Neural Network that Explains Its Predictionscited by 234
- 2020Neural Prototype Trees for Interpretable Fine-grained Image Recognitioncited by 213
- 2021How to represent part-whole hierarchies in a neural networkcited by 123
- 2018Robust Classification with Convolutional Prototype Learningcited by 122
- 2021Deformable ProtoPNet: An Interpretable Image Classifier Using Deformable Prototypescited by 120
- 2017DeepIGeoS: A Deep Interactive Geodesic Framework for Medical Image Segmentationcited by 110
- 2019Interpretable Image Recognition with Hierarchical Prototypescited by 64
- 2021Prototypical Cross-Attention Networks for Multiple Object Tracking and Segmentationcited by 43
Built on by
Citing papers per year: 2025: 1.
1 paper.