# Achieving Linear Speedup and Near-Optimal Complexity for Decentralized Optimization over Row-stochastic Networks

arXiv: https://arxiv.org/abs/2506.04600 · HTML: https://mlsanity.com/paper/2506.04600/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Jun 2025.

## Built on by (most cited first, 2 of 2)

- [Near-Optimal Decentralized Stochastic Nonconvex Optimization with Heavy-Tailed Noise](https://mlsanity.com/paper/2601.11435/) (2026)
- [Row-Stochastic Matrices Can Provably Outperform Doubly Stochastic Matrices in Decentralized Learning](https://mlsanity.com/paper/2511.19513/) (2025)

## Abstract (arXiv)

A key challenge in decentralized optimization is determining the optimal convergence rate and designing algorithms to achieve it. While this problem has been extensively addressed for doubly-stochastic and column-stochastic mixing matrices, the row-stochastic scenario remains unexplored. This paper bridges this gap by introducing effective metrics to capture the influence of row-stochastic mixing matrices and establishing the first convergence lower bound for decentralized learning over row-stochastic networks. However, existing algorithms fail to attain this lower bound due to two key issues: deviation in the descent direction caused by the adapted gradient tracking (GT) and instability introduced by the Pull-Diag protocol. To address descent deviation, we propose a novel analysis framework demonstrating that Pull-Diag-GT achieves linear speedup, the first such result for row-stochastic decentralized optimization. Moreover, by incorporating a multi-step gossip (MG) protocol, we resolve the instability issue and attain the lower bound, achieving near-optimal complexity for decentralized optimization over row-stochastic networks.