T-SKM-Net: Trainable Neural Network Framework for Linear Constraint Satisfaction via Sampling Kaczmarz-Motzkin Method
Published in Poster, AAAI-26 (scheduled for February 2026), 2025
AAAI-26 poster: integrates Sampling Kaczmarz-Motzkin into a trainable neural network framework for fast, constraint-satisfying inference.
Recommended citation: Zhu, Haoyu; Zhang, Yao; Ren, Jiashen; and Hou, Qingchun. 2025. "T-SKM-Net: Trainable Neural Network Framework for Linear Constraint Satisfaction via Sampling Kaczmarz-Motzkin Method." arXiv preprint arXiv:2512.10461. Poster to appear at AAAI-26.
Download Paper