Federated Learning over Wireless Networks: A Band-limited Coordinated Descent Approach

Abstract

We consider a many-to-one wireless architecture for federated learning at the network edge, where multiple edge devices collaboratively train a model using local data. The unreliable nature of wireless connectivity, together with constraints in computing resources at edge devices, dictates that the local updates at edge devices should be carefully crafted and compressed to match the wireless communication resources available and should work in concert with the receiver. Thus motivated, we propose SGD-based bandlimited coordinate descent algorithms for such settings. Specifically, for the wireless edge employing over-the-air computing, a common subset of k-coordinates of the gradient updates across edge devices are selected by the receiver in each iteration, and then transmitted simultaneously over k sub-carriers, each experiencing time-varying channel conditions. We characterize the impact of communication error and compression, in terms of the resulting gradient bias and mean squared error, on the convergence of the proposed algorithms. We then study learning-driven communication error minimization via joint optimization of power allocation and learning rates. Our findings reveal that optimal power allocation across different sub-carriers should take into account both the gradient values and channel conditions, thus generalizing the widely used water-filling policy. We also develop sub-optimal distributed solutions amenable to implementation.

References in corpus (12)

• Deep Residual Learning for Image Recognition
• Federated Learning: Strategies for Improving Communication Efficiency
• QSGD: Communication-Efficient SGD via Gradient Quantization and Encoding
• Machine Learning at the Wireless Edge: Distributed Stochastic Gradient Descent Over-the-Air
• Sparsified SGD with Memory
• On Analog Gradient Descent Learning over Multiple Access Fading Channels
• Over-the-air Function Computation in Sensor Networks
• Towards an Intelligent Edge: Wireless Communication Meets Machine Learning
• Federated Learning via Over-the-Air Computation
• Energy-Efficient Radio Resource Allocation for Federated Edge Learning
• Communication-efficient distributed SGD with Sketching
• Blind Over-the-Air Computation and Data Fusion via Provable Wirtinger Flow

Cited by in corpus (1)

• Optimized Power Control Design for Over-the-Air Federated Edge Learning