Classifications are not very informative

“The problem with Bernoulli regression is that binary outcomes just aren’t very informative,” one of my colleagues said to me in the context of a regression problem. Now I have decided that there is some meat on this bone. 🚧TODO🚧 clarify the informativeness of categories about their covariates for the post-Imagenet era, from a classic vector quantization perspective.

Then: Deep learning classifiers as a model for legibility.

TODO: write up the beautiful worked example from psychometrics. The Ultimate Personality Test 2.0