Federated Learning: From Theory to Practice

Federated Learning: From Theory to Practice

Alexander Jung · Springer, 2026
ISBN: 978-981-95-1008-5 (Hardcover), 978-981-95-1011-5 (Softcover), 978-981-95-1009-2 (eBook)

Federated Learning: From Theory to Practice develops federated learning systems from one flexible design principle: generalized total variation minimization (GTVMin), the natural analogue of the empirical risk minimization that underpins classical machine learning. Devices form a federated learning network whose edges encode communication links and task similarity, and training personalized models becomes a distributed optimization problem over this network. The book builds directly on Machine Learning: The Basics.

Get the book

Cite

@book{Jung2026FL,
  author    = {Jung, Alexander},
  title     = {Federated Learning: From Theory to Practice},
  publisher = {Springer},
  address   = {Singapore},
  year      = {2026},
  isbn      = {978-981-95-1009-2}
}