… Abstract—In this paper, the problem of training federated learning (FL) algorithms over a
realistic wireless network is studied. In particular, in the considered model, wireless users …

… One such distributed ML architecture is federated learning (FL), which allows a decoupling
of data provision at end-user equipment (UE) and machine learning model aggregation at a …

… We refer to this simply as federated optimization, although this is … performance comparison
of three federated learning algorithms, and (2) comparing them to regular centralized learning…

… Learning time and UE energy consumption. We fill this gap by formulating a Federated
Learning over wireless network as an optimization problem FEDL that captures both trade-offs. …

… The works in [6] and [7] optimized the FL performance with … evaluate the FL performance and
optimized the batchsize selection … In [15], the authors developed a federated learning based …

… Federated Learning Let us consider a general FL system consisting of a honestbut-curious
server and U … Convergence for Private Federated Learning In this subsection, we analyze the …

… communication and computation efficiency in federated learning (FL) via model quantization…
We pose this as an optimization problem that aims to minimize the training loss of quantized …

… Abstract—Federated learning (FL) is a distributed learning … Literature in FL cost optimization
mainly focused on learning … learning time was studied in [25]–[32], and joint optimization …

… Co-optimizations of statistical and system efficiency Most of today’s FL efforts focus on either
… , whereas we observe a great opportunity to jointly optimize both efficiencies: (1) As the …

… learning method to process federated training. FC API, the basic layer for federation learning…
In this section, this study discusses about the evolution and optimization in the following. We …