Learning Fair and Interpretable Representations via Linear Orthogonalization

Open in new window