Get the latest tech news
Toy Models of Superposition (2022)
It would be very convenient if the individual neurons of artificial neural networks corresponded to cleanly interpretable features of the input. For example, in an “ideal” ImageNet classifier, each neuron would fire only in the presence of a specific visual feature, such as the color red, a left-facing curve, or a dog snout.
A number of previous interpretability papers have considered it , and it's very closely related to the long-studied topic of compressed sensing in mathematics , as well as the ideas of distributed, dense, and population codes in neuroscience and deep learning . In the previous diagram, we found that there are distinct lines corresponding to dimensionality of: ¾ (tetrahedron), ⅔ (triangle), ½ (antipodal pair), ⅖ (pentagon), ⅜ (square antiprism), and 0 (feature not learned). Concretely, disentanglement research often explores whether one can train a VAE or GAN where basis dimensions correspond to the major features one might use to describe the problem (e.g. rotation, lighting, gender… as relevant).
Or read this on Hacker News