paper-with-me

홈 › Papers

Importance of Smoothness Induced by Optimizers in FL4ASR: Towards Understanding Federated Learning for End-to-End ASR

2023-09-22 · Sheikh Shams Azam, Tatiana Likhomanenko, Martin Pelikan, Jan "Honza" Silovsky

In this paper, we start by training End-to-End Automatic Speech Recognition (ASR) models using Federated Learning (FL) and examining the fundamental considerations that can be pivotal in minimizing the performance gap in terms of word error rate between models trained using FL versus their centralized counterpart. Specifically, we study the effect of (i) adaptive optimizers, (ii) loss characteristics via altering Connectionist Temporal Classification (CTC) weight, (iii) model initialization through seed start, (iv) carrying over modeling setup from experiences in centralized training to FL, e.g., pre-layer or post-layer normalization, and (v) FL-specific hyperparameters, such as number of local epochs, client sampling size, and learning rate scheduler, specifically for ASR under heterogeneous data distribution. We shed light on how some optimizers work better than others via inducing smoothness. We also summarize the applicability of algorithms, trends, and propose best practices from prior works in FL (in general) toward End-to-End ASR models.

📄 PDF Abstract BibTeX arXiv:2309.13102

Code (0)

등록된 구현이 없습니다.

Tasks

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Federated Learningspeech-recognitionSpeech Recognition

Similar Papers 제목 키워드 기반

On the Width Scaling of Neural Optimizers Under Matrix Operator Norms I: Row/Column Normalization and Hyperparameter Transfer

2026-03-10 · Ruihan Xu, Jiajin Li, Yiping Lu arxiv

A central question in modern deep learning is how to design optimizers whose behavior remains stable as the network width $w$ increases. We address this question by interpreting several widely used neural-network optimiz…

A Tale of Two Geometries: Adaptive Optimizers and Non-Euclidean Descent

2025-11-25 · Shuo Xie, Tianhao Wang, Beining Wu, Zhiyuan Li arxiv

Adaptive optimizers can reduce to normalized steepest descent (NSD) when only adapting to the current gradient, suggesting a close connection between the two algorithmic families. A key distinction between their analyses…

Stochastic Optimization

Adaptive Federated Optimization

2020-02-29 · ICLR 2021 1 · Sashank Reddi, Zachary Charles, Manzil Zaheer, Zachary Garrett 외

Federated learning is a distributed machine learning paradigm in which a large number of clients coordinate with a central server to learn a model without sharing their own training data. Standard federated optimization …

Federated Learning

Technical note on Fisher Information for Robust Federated Cross-Validation

2025-10-04 · Behraj Khan, Tahir Qasim Syed arxiv

When training data are fragmented across batches or federated-learned across different geographic locations, trained models manifest performance degradation. That degradation partly owes to covariate shift induced by dat…

Federated Learning

A Convergence Theory for Federated Average: Beyond Smoothness

2022-11-03 · Xiaoxiao Li, Zhao Song, Runzhou Tao, Guangyi Zhang

Federated learning enables a large amount of edge computing devices to learn a model without data sharing jointly. As a leading algorithm in this setting, Federated Average FedAvg, which runs Stochastic Gradient Descent …

Edge-computingFederated Learning