# CA-HFP: Curvature-Aware Heterogeneous Federated Pruning with Model Reconstruction

arXiv: https://arxiv.org/abs/2603.12591 · HTML: https://mlsanity.com/paper/2603.12591/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 13 references that are in the map.
- First posted on arXiv 13 Mar 2026.

## Builds on (most cited first, 10 of 10)

- [Communication-Efficient Learning of Deep Networks from Decentralized Data](https://mlsanity.com/paper/1602.05629/) (2016)
- [Advances and Open Problems in Federated Learning](https://mlsanity.com/paper/1912.04977/) (2019)
- [Federated Optimization in Heterogeneous Networks](https://mlsanity.com/paper/1812.06127/) (2018)
- [Federated Learning in Mobile Edge Networks: A Comprehensive Survey](https://mlsanity.com/paper/1909.11875/) (2019)
- [Model Pruning Enables Efficient Federated Learning on Edge Devices](https://mlsanity.com/paper/1909.12326/) (2019)
- [Federated Learning and Wireless Communications](https://mlsanity.com/paper/2005.05265/) (2020)
- [Complement Sparsification: Low-Overhead Model Pruning for Federated Learning](https://mlsanity.com/paper/2303.06237/) (2023)
- [DapperFL: Domain Adaptive Federated Learning with Model Fusion Pruning for Edge Devices](https://mlsanity.com/paper/2412.05823/) (2024)
- [FedLPS: Heterogeneous Federated Learning for Multiple Tasks with Local Parameter Sharing](https://mlsanity.com/paper/2402.08578/) (2024)
- [Efficient Federated Learning with Enhanced Privacy via Lottery Ticket Pruning in Edge Computing](https://mlsanity.com/paper/2305.01387/) (2023)

## Built on by (most cited first, 1 of 1)

- [FedSAP: Federated Learning with Structured Adaptive Partitioning for Multi-Domain Heterogeneous Edge Devices](https://mlsanity.com/paper/2610.01638/) (2026)

## Abstract (arXiv)

Federated learning on heterogeneous edge devices requires personalized compression while preserving aggregation compatibility and stable convergence. We present Curvature-Aware Heterogeneous Federated Pruning (CA-HFP), a practical framework that enables each client perform structured, device-specific pruning guided by a curvature-informed significance score, and subsequently maps its compact submodel back into a common global parameter space via a lightweight reconstruction. We derive a convergence bound for federated optimization with multiple local SGD steps that explicitly accounts for local computation, data heterogeneity, and pruning-induced perturbations; from which a principled loss-based pruning criterion is derived. Extensive experiments on FMNIST, CIFAR-10, and CIFAR-100 using VGG and ResNet architectures under varying degrees of data heterogeneity demonstrate that CA-HFP preserves model accuracy while significantly reducing per-client computation and communication costs, outperforming standard federated training and existing pruning-based baselines.