Bartosz Milewski on Nostr: nprofile1q…fv2av Good point. If the matrix W_V were invertible, than we could ...

nprofile1qy2hwumn8ghj7un9d3shjtnddaehgu3wwp6kyqpq73zugfcw5qemam49j598gfhr5nyctxun4mt3y8qf4ayj4pdnt7gqkfv2av (nprofile…v2av) Good point. If the matrix W_V were invertible, than we could absorb it into the definition of E's and the two other matrices. My guess is that all these matrices project onto smaller-dimensional subspaces, thus cannot be inverted. But I don't know how much of it is just optimization heuristics.