K Deng, J Hu - arXiv preprint arXiv:2304.08241, 2023 - arxiv.org
We consider the problem of decentralized nonconvex optimization over a compact submanifold, where each local agent's objective function defined by the local dataset is …
L Wang, X Liu - IEEE Transactions on Signal Processing, 2022 - ieeexplore.ieee.org
In this paper, we focus on the decentralized optimization problem over the Stiefel manifold, which is defined on a connected network of agents. The objective is an average of local …
The conjugate gradient method is a crucial first-order optimization method that generally converges faster than the steepest descent method, and its computational cost is much …
H Ye, T Zhang - Journal of Machine Learning Research, 2021 - jmlr.org
Due to the rapid growth of smart agents such as weakly connected computational nodes and sensors, developing decentralized algorithms that can perform computations on local agents …
In this paper, we tackle a significant challenge in PCA: heterogeneity. When data are collected from different sources with heterogeneous trends while still sharing some …
J Li, S Ma - arXiv preprint arXiv:2206.05668, 2022 - arxiv.org
Federated learning (FL) has found many important applications in smart-phone-APP based machine learning applications. Although many algorithms have been studied for FL, to the …
We focus on a class of non-smooth optimization problems over the Stiefel manifold in the decentralized setting, where a connected network of $ n $ agents cooperatively minimize a …
A Gang, B Xiang, WU Bajwa - IEEE Transactions on Signal and …, 2021 - ieeexplore.ieee.org
Principal Subspace Analysis (PSA)—and its sibling, Principal Component Analysis (PCA)— is one of the most popular approaches for dimensionality reduction in signal processing and …
X Wang, Y Jiao, HT Wai, Y Gu - International Conference on …, 2023 - proceedings.mlr.press
We consider the problem of distributed principal component analysis (PCA) where the data samples are dispersed across different agents. Despite the rich literature on this problem …