Explaining neural networks
Explaining neural networks
Explaining neural networks - Day 14 lecture of the Foundations of Machine Learning in Python course.
High-Performance Computing and Analytics Lab, University of Bonn
Bonn
Topics covered in this lesson
- Linear classifiers
- Interpretation by examination
- Case Study: Deepfake detection
- Input Optimization
- The problem with deep CNN
- Saliency Maps and Integrated Gradients
- Literature
Documents
Day 14 Explaining neural networks
(6.35 MB)
Day 14 Exercise_interpretability.zip
(34.08 MB)