Deploying federated learning across heterogeneous IoT device fleets requires tailored neural network architectures for each device class, yet existing Federated Neural Architecture Search (FedNAS) methods suffer from unguided supernet training and prohibitively costly post-training search pipelines that demand over 20 GPU-hours per deployment target. We introduce DeepFedNAS, a two-phase framework built on a multi-objective fitness function that synthesizes information-theoretic network metrics with architectural heuristics. In the first phase, Federated Pareto Optimal Supernet Training replaces random subnet sampling with a pre-computed cache of elite, high-fitness architectures, yielding a superior supernet. In the second phase, a Predictor-Free Search uses this fitness function as a zero-cost accuracy proxy, discovering hardware-optimized subnets in ~20 seconds, a ~61x speedup over the baseline pipeline. Experiments on CIFAR-10, CIFAR-100, and CINIC-10 demonstrate state-of-the-art accuracy (up to +1.21% on CIFAR-100), a 2.8x reduction in per-round transmission size, and robust performance under extreme non-IID conditions (α = 0.1), making DeepFedNAS practical for scalable, communication-constrained IoT federations. Source code: https://github.com/bostankhan6/DeepFedNAS

翻译：在异构物联网设备群上部署联邦学习需要为每类设备定制神经网络架构，然而现有联邦神经架构搜索（FedNAS）方法存在超网络训练缺乏引导、训练后搜索流水线成本过高（每个部署目标需消耗超过20 GPU小时）的问题。我们提出DeepFedNAS，一种基于多目标适应度函数的两阶段框架，该函数将信息论网络度量与架构启发式策略相结合。第一阶段中，联邦帕累托最优超网络训练以预计算的精英高适应度架构缓存替代随机子网络采样，生成更优的超网络。第二阶段中，无预测器搜索将该适应度函数作为零成本精度代理，可在约20秒内发现硬件优化的子网络，相较基准流水线实现约61倍加速。在CIFAR-10、CIFAR-100和CINIC-10上的实验表明，该方法达到了最先进的准确率（CIFAR-100上提升高达+1.21%），每轮传输大小减少2.8倍，并在极端非独立同分布条件（α=0.1）下保持稳健性能，使DeepFedNAS适用于可扩展、通信受限的物联网联邦场景。源代码：https://github.com/bostankhan6/DeepFedNAS