Discovering Sparse Representations of Lie Groups with Machine Learning

Recent work has used deep learning to derive symmetry transformations, which preserve conserved quantities, and obtain the corresponding algebras of generators. In this letter, we extend technique sparse representations arbitrary Lie algebras. We show that our method reproduces canonical (sparse) generators Lorentz group, as well U(n) SU(n) families groups. This approach is completely general can be find infinitesimal for any group.