Completed
Applications of Deep Learning
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Understanding Deep Neural Networks - CVPR 2020 iMLCV Tutorial
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Applications of Deep Learning
- 3 Research Themes
- 4 Prior Work: Perturbation Approaches
- 5 Our Approach: Meaningful Perturbations
- 6 Our Approach: Extremal Perturbations
- 7 Interpretability
- 8 Foreground evidence is usually sufficient
- 9 Suppressing the background may overdrive the network
- 10 Adversarial Defense
- 11 Regularization to mitigate artifacts
- 12 Area Constraint
- 13 Smooth Masks
- 14 Comparison with Prior Work
- 15 Measure Performance on Weak Localization
- 16 Selectivity to Output Class
- 17 Sensitive to Model Parameters
- 18 Intermediate Activations
- 19 Spatial Attribution
- 20 Channel Attribution
- 21 Activation "Diffing"
- 22 # Concepts per Filter
- 23 # Filters per Concept
- 24 Self-Supervised Learning
- 25 Comparing Concept Embedding Spaces
- 26 Segmentation
- 27 Classification
- 28 Human-Guided Machine Learning
- 29 Future Work: Model Debugging
