AFAFed -- Protocol analysis

TL;DR

AFAFed is an innovative asynchronous federated learning framework designed for IoT environments, featuring adaptive mechanisms for fairness and communication efficiency, with proven convergence even under non-convex conditions.

Contribution

The paper introduces AFAFed, a novel federated learning protocol with adaptive fairness and communication tuning, along with new convergence analysis for non-convex loss functions.

Findings

Convergence guarantees under non-convex loss functions.

Impact of FL parameters on convergence rate.

Effective adaptation to heterogeneous and unreliable IoT environments.

Abstract

In this paper, we design, analyze the convergence properties and address the implementation aspects of AFAFed. This is a novel Asynchronous Fair Adaptive Federated learning framework for stream-oriented IoT application environments, which are featured by time-varying operating conditions, heterogeneous resource-limited devices (i.e., coworkers), non-i.i.d. local training data and unreliable communication links. The key new of AFAFed is the synergic co-design of: (i) two sets of adaptively tuned tolerance thresholds and fairness coefficients at the coworkers and central server, respectively; and, (ii) a distributed adaptive mechanism, which allows each coworker to adaptively tune own communication rate. The convergence properties of AFAFed under (possibly) non-convex loss functions is guaranteed by a set of new analytical bounds, which formally unveil the impact on the resulting AFAFed convergence rate of a number of Federated Learning (FL) parameters, like, first and second moments of the per-coworker number of consecutive model updates, data skewness, communication packet-loss probability, and maximum/minimum values of the (adaptively tuned) mixing coefficient used for model aggregation.