no code implementations • 26 Sep 2023 • Yahong Yang, Qipin Chen, Wenrui Hao
In this paper, we present a novel training approach called the Homotopy Relaxation Training Algorithm (HRTA), aimed at accelerating the training process in contrast to traditional methods.
no code implementations • 1 Sep 2021 • Qipin Chen, Wenrui Hao, Juncai He
To address this challenge, we study neural networks from a nonlinear computation point of view and propose a novel weight initialization strategy that is based on the linear product structure (LPS) of neural networks.
no code implementations • ICLR 2020 • Qipin Chen, Wenrui Hao
We present a Homotopy Training Algorithm (HTA) to solve optimization problems arising from neural networks.
