This code is released for the implementation of ADMM for the training of deep sigmoid nets.
This is a demo to show how to use ADMM proposed in [1] to train a deep sigmoid nets in two approximation tasks. Please cite the following reference:
[1] Jinshan Zeng, Shao-Bo Lin, Yuan Yao and Ding-Xuan Zhou, On ADMM in deep learning: Convergence and saturation-avoidance, Journal of Machine Learning Research, 22(199): 1-67, 2021.