Stochastic gradient descent with model building

[Neurips 2022] "Deep Architecture Connectivity Matters for Its Convergence: A Fine-Grained Analysis" by Wuyang Chen*, Wei Huang*, Xinyu Gong, Boris Hanin, Zhangyang Wang

Worth-reading papers and related awesome resources on deep learning optimization algorithms. 值得一读的深度学习优化器论文与相关资源。

Alpha Belief Propagation

# warp-convergence-analysis

Special Project - CA classification (2019 Fall)

This repository contains code for analyzing the convergence, inference, and regret of Stream SGD, as presented in our paper. It includes implementations for queueing systems and inventory control.

Estimating π using multiple numerical algorithms, with clean architecture, reproducibility, and scientific visualization.

Bayesian inference tools, allowing to execute Gibbs sampling and run some diagnostics on results.

This repository is for reproducing the experimental results in our manuscript submitted to IEEE TIT. An early version of this work have been presented in part at ICML'24.

Analysis of simulation data obtained with the ParMooN finite element package.

Evaluating several new approaches to improve convergence of Randomized Kaczmarz (RK) for consistent ill-conditioned systems. This project explores the availability of convergence information among pairwise row differences and analyzes sampling techniques that involve clustering and spectral analysis.