Tensor rank and the ill-posedness of the best low-rank approximation problem

Probably useful for my 🗺️ PhD Plan to reduce the size of the Neural networks used.

• Look into the Eckart–Young theorem.