Myosotis: structured computation for attention like layer

Attention layers apply a sequence-to-sequence mapping whose parameters depend on the pairwise interactions of the input elements.However, without any structural assumptions, the memory and compute scale quadratically with the sequence length. The two main ways to mitigate this are to introduce sparsity by ignoring a sufficient amount of pairwise interactions or to introduce recurrent dependence along them, as SSM does. Although both approaches are reasonable, they both have disadvantages. We propose a novel algorithm that combines the advantages of both concepts. Our idea is based on the efficient inversion of tree-structured matrices.