Micro Blossom:面向量子纠错的加速最小权重完美匹配解码器
计算机视觉与模式识别
2025-02-21 v1 人工智能
摘要
最小权重完美匹配 (MWPM) 解码器是量子纠错解码中重要方法,因其准确率受到广泛关注。然而,许多人认为实现超导量子比特提出的微秒级延迟要求几乎不可能。本工作提出首个公开知晓的MWPM解码器,名为Micro Blossom,实现亚微秒级解码延迟。Micro Blossom采用异构架构,将最先进的MWPM解码器在软件和可编程加速器之间进行精细划分,加速器拥有与解码图每个顶点/边各一个的平行处理单元。在代码距离为d、采用电路级噪声模型且物理错误率为p的情形下,Micro Blossom的加速器采用O(d^3)个平行处理单元,将最坏情况延迟从O(d^12)降至O(d^9),将平均延迟从O(pd^3+1)降至O(p^2d^2+1)(当p≪1时)。我们报告了使用FPGA实现的原型系统。在d=13且p=0.1%时,原型实现的平均解码延迟为0.8微秒,时钟频率为62 MHz。Micro Blossom是首个公开知晓的硬件加速精确MWPM解码器,其0.8微秒解码延迟是已报道文献中最佳MWPM解码器实现的最短延迟的8倍。
引用
@article{arxiv.2502.14786,
title = {SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features},
author = {Michael Tschannen and Alexey Gritsenko and Xiao Wang and Muhammad Ferjad Naeem and Ibrahim Alabdulmohsin and Nikhil Parthasarathy and Talfan Evans and Lucas Beyer and Ye Xia and Basil Mustafa and Olivier Hénaff and Jeremiah Harmsen and Andreas Steiner and Xiaohua Zhai},
journal= {arXiv preprint arXiv:2502.14786},
year = {2025}
}
备注
Model checkpoints are available at https://github.com/google-research/big_vision/tree/main/big_vision/configs/proj/image_text/README_siglip2.md