1 paper across 1 session
Equivariant neural networks can fail to approximate target functions even when they achieve maximal separation power.
