Pixel-grounded prototypical part networks
Prototypical part neural networks (ProtoPartNNs), namely ProtoPNet and its derivatives, are
an intrinsically interpretable approach to machine learning. Their prototype learning scheme …
an intrinsically interpretable approach to machine learning. Their prototype learning scheme …
This Probably Looks Exactly Like That: An Invertible Prototypical Network
We combine concept-based neural networks with generative, flow-based classifiers into a
novel, intrinsically explainable, exactly invertible approach to supervised learning …
novel, intrinsically explainable, exactly invertible approach to supervised learning …
LucidPPN: Unambiguous Prototypical Parts Network for User-centric Interpretable Computer Vision
M Pach, D Rymarczyk, K Lewandowska… - arXiv preprint arXiv …, 2024 - arxiv.org
Prototypical parts networks combine the power of deep learning with the explainability of
case-based reasoning to make accurate, interpretable decisions. They follow the this looks …
case-based reasoning to make accurate, interpretable decisions. They follow the this looks …
Revisiting FunnyBirds evaluation framework for prototypical parts networks
S Opłatek, D Rymarczyk, B Zieliński - World Conference on Explainable …, 2024 - Springer
Prototypical parts networks, such as ProtoPNet, became popular due to their potential to
produce more genuine explanations than post-hoc methods. However, for a long time, this …
produce more genuine explanations than post-hoc methods. However, for a long time, this …
Prototype-based Interpretable Breast Cancer Prediction Models: Analysis and Challenges
S Pathak, J Schlötterer, J Veltman, J Geerdink… - World Conference on …, 2024 - Springer
Deep learning models have achieved high performance in medical applications, however,
their adoption in clinical practice is hindered due to their black-box nature. Using …
their adoption in clinical practice is hindered due to their black-box nature. Using …
ProtoS-ViT: Visual foundation models for sparse self-explainable classifications
Prototypical networks aim to build intrinsically explainable models based on the linear
summation of concepts. However, important challenges remain in the transparency …
summation of concepts. However, important challenges remain in the transparency …
[PDF][PDF] Explainable AI for High-stakes Decision-making
Z Carmichael - 2024 - curate.nd.edu
As a result of the many recent advancements in artificial intelligence (AI), a significant
interest in the technology has developed from high-stakes decision-makers in industries …
interest in the technology has developed from high-stakes decision-makers in industries …