Understanding Visual Feature Reliance through the Lens of Complexity
Paper
•
2407.06076
•
Published
•
7
Thomas FEL
ThomasFEL
AI & ML interests
Explainability, Computer Vision, Optim, Alignment
Organizations
authored
7
papers
over 1 year ago
CRAFT: Concept Recursive Activation FacTorization for Explainability
Paper
•
2211.10154
•
Published
Don't Lie to Me! Robust and Efficient Explainability with Verified Perturbation Analysis
Paper
•
2202.07728
•
Published
A Holistic Approach to Unifying Automatic Concept Extraction and Concept Importance Estimation
Paper
•
2306.07304
•
Published
Unlocking Feature Visualization for Deeper Networks with MAgnitude Constrained Optimization
Paper
•
2306.06805
•
Published
Harmonizing the object recognition strategies of deep neural networks with humans
Paper
•
2211.04533
•
Published
Xplique: A Deep Learning Explainability Toolbox
Paper
•
2206.04394
•
Published