Introducton to Circuits in Convolutional Neural Networks
Continuing our running theme this week on interpretability in deep learning neural networks, let's check out a tutorial talk by Chris Olah titled Introduction to Circuits in CNNs. This is from a workshop at CVPR 2020.
If you are have read any of the Distil publications on feature visualization, you will find this talk interesting.
1: Complex Gabor filters. In reciprocal pairs.
2: Circle detector made from curve detectors.
3. Triangle detector made from line detectors. Really detecting 'inflection points'.
4: Absence of color vs color contrast detectors.
5: Boundary detectors. Meta level higher order thing compared to normal edge detector. High-low frequency boundary, etc.
6: Clean room implementation of curve detector.