A Hierarchical Sampling Framework for bounding the Generalization Error of Federated Learning

2026-05-05 · Dario Filatrella, Ragnar Thobaben, Mikael Skoglund arxiv

We study expected generalization bounds for the Hierarchical Federated Learning (HFL) setup using Wasserstein distance. We introduce a generalized framework in which data is sampled hierarchically, and we model it with a multi-layered tree structure that induces dependencies among the clients' datasets. We derive generalization bounds in terms of Wasserstein distance under the Lipschitz assumption on the loss function, by applying a supersample construction that allows us to measure the sensitivity of the algorithm to the change of a single node in the sampling tree. By leveraging the FL structure, we recover and strictly imply existing state-of-the-art conditional mutual information (CMI) bounds in the case of bounded losses. We also show that our bound can be applied together with Differential Privacy assumptions, to recover generalization bounds based on algorithmic privacy. To assess the tightness of our bounds, we study the Gaussian Location Model (GLM) and show that we recover the actual asymptotic rate of the generalization error.

📄 PDF Abstract BibTeX arXiv:2605.03499

Code (0)

등록된 구현이 없습니다.

Tasks

Federated Learning

A Look at the Effect of Sample Design on Generalization through the Lens of Spectral Analysis

2019-06-06 · Bhavya Kailkhura, Jayaraman J. Thiagarajan, Qunwei Li, Peer-Timo Bremer

This paper provides a general framework to study the effect of sampling properties of training data on the generalization error of the learned machine learning (ML) models. Specifically, we propose a new spectral analysi…

valid

Generalization Bounds and Statistical Guarantees for Multi-Task and Multiple Operator Learning with MNO Networks

2026-04-02 · Adrien Weihs, Hayden Schaeffer arxiv

Multiple operator learning concerns learning operator families $\{G[α]:U\to V\}_{α\in W}$ indexed by an operator descriptor $α$. Training data are collected hierarchically by sampling operator instances $α$, then input f…

Generalization Bounds for Graph Embedding Using Negative Sampling: Linear vs Hyperbolic

2021-12-01 · NeurIPS 2021 12 · Atsushi Suzuki, Atsushi Nitanda, Jing Wang, Linchuan Xu 외

Graph embedding, which represents real-world entities in a mathematical space, has enabled numerous applications such as analyzing natural languages, social networks, biochemical networks, and knowledge bases.It has been…

Generalization Bounds Graph Embedding

Two-Layer Generalization Analysis for Ranking Using Rademacher Average

2010-12-01 · NeurIPS 2010 12 · Wei Chen, Tie-Yan Liu, Zhi-Ming Ma

This paper is concerned with the generalization analysis on learning to rank for information retrieval (IR). In IR, data are hierarchically organized, i.e., consisting of queries and documents per query. Previous general…

Generalization Bounds Information Retrieval Learning-To-Rank Retrieval+1

Exploring Active 3D Object Detection from a Generalization Perspective

2023-01-23 · Yadan Luo, Zhuoxiao Chen, Zijian Wang, Xin Yu 외

To alleviate the high annotation cost in LiDAR-based 3D object detection, active learning is a promising solution that learns to select only a small portion of unlabeled data to annotate, without compromising model perfo…

3D Object Detection Active Learning Informativeness object-detection+1