Doubly Robust Fusion of Many Treatments for Policy Learning | Read Paper on Bytez