that is absolutely fascinating. curious to read the paper, and see what the major obstacles were, or whether some newer techniques (like federated learning) cou

that is absolutely fascinating. curious to read the paper, and see what the major obstacles were, or whether some newer techniques (like federated learning) could make the process more computationally efficient