A thread on my experience with misogyny in academic science.

First some background: Reading @kate_manne helped me understand misogyny as the ‘law enforcement’ branch of the patriarchy. It’s effects are felt most strongly by those who violate gender norms. I’ve seen this play out in research where certain men have successful professional relationships with women when the woman is clearly in the role of the mentee. If she plays the role of the junior researcher/student who needs some expertise that he has, all may go smoothly.

WIP: What happens to intermediate representations as more layers are left at their random initialization? Here: the similarity between a convnet trained normally (all layers trained) and networks where only layers after layer n are trained (early layers untrained). Notice that the upper right quadrant remains roughly the same, even as the number of trained layers shrinks. Largest changes in off-diagonal similarity seems to occur below layer cnn7. This boundary may reflect bottom-up vs top-town information propagation?