activity
20182022
most citedA Convergent ADMM Framework for Efficient Neural Network Training

2 citations · 4 across the 4 of their papers we have counts for

collaborators

5 papers

cs.IT2022★ 1 cited

Edge Graph Neural Networks for Massive MIMO Detection

Hongyi Li, Junxiang Wang, Yongchao Wang

Massive Multiple-Input Multiple-Out (MIMO) detection is an important problem in modern wireless communication systems. While traditional Belief Propagation (BP) detectors perform p…

cs.LG2021★ 2 cited

A Convergent ADMM Framework for Efficient Neural Network Training

Junxiang Wang, Hongyi Li, Liang Zhao

As a well-known optimization framework, the Alternating Direction Method of Multipliers (ADMM) has achieved tremendous success in many classification and regression applications. R…

cs.LG2021★ 1 cited

Community-based Layerwise Distributed Training of Graph Convolutional Networks

Hongyi Li, Junxiang Wang, Yongchao Wang +2

The Graph Convolutional Network (GCN) has been successfully applied to many graph-based applications. Training a large-scale GCN model, however, is still challenging: Due to the no…

cs.LG2021

Towards Quantized Model Parallelism for Graph-Augmented MLPs Based on Gradient-Free ADMM Framework

Junxiang Wang, Hongyi Li, Zheng Chai +3

While Graph Neural Networks (GNNs) are popular in the deep learning community, they suffer from several challenges including over-smoothing, over-squashing, and gradient vanishing.…

math.OC2018

Accelerated Gradient-free Neural Network Training by Multi-convex Alternating Optimization

Junxiang Wang, Hongyi Li, Liang Zhao

In recent years, even though Stochastic Gradient Descent (SGD) and its variants are well-known for training neural networks, it suffers from limitations such as the lack of theoret…