Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Story

GPU solver cuLoRADS cracks 18-year-old quantum ordered-search problem

AI summary

A Shanghai Jiao Tong University team used the GPU-accelerated solver cuLoRADS to directly solve the full semidefinite program for five quantum ordered-search queries, confirming that the maximum searchable ordered-list length is 7,265 and that the scheme fails at 7,266, according to MIT Technology Review China citing a PNAS paper. The program had grown quadratically with list length for 18 years, beyond the reach of CPU solvers at the time. The paper reports that on a single NVIDIA H100, a set of 10-million-order MaxCut semidefinite relaxations was solved in 10 seconds to one minute, versus tens of hours for previously reported CPU methods.

In-depth insight

cuLoRADS provides a directly usable optimization tool for quantum, quantum-chemistry and embodied-AI research, illustrating algorithm-hardware co-design for scientific computing.

一个在量子信息领域悬了 18 年的难解之题,竟然被一张 GPU 解开了。这个问题叫做量子有序搜索,属于量子算法复杂度研究领域的一个基础问题。近日,上海交通大学团队用 GPU 加速的求解器 cuLoRADS 直接解出了 5 次查询对应的完整半定规划(SDP,Semidefinite Programming),确认了最大可搜索列表规模是 7265,相关论文发表在 PNAS 上。

过去 18 年间,相关半定规划问题的规模随着搜索列表长度呈现出平方级增长,这超出了所有 CPU 求解器的能力范围。

半定规划是一类数学优化方法,在组合优化、机器学习、控制和量子科学里都有一定的应用。它的建模能力非常强,计算代价也很高。规模一旦变大,传统方法就会解不动。cuLoRADS 则把这类问题的求解搬到了 GPU 上,同时根据 GPU 的特点重新设计了核心计算流程。

打个比方来说,传统 CPU 求解器在处理半定规划的时候,就像一个班主任逐条核对成绩表,一条一条地来,不仅速度慢,内存也非常吃紧。cuLoRADS 则是把成绩单摊开,让 GPU 上的成千上万个小核心在同一时间开工,各自去算一部分。论文数据显示,在单张英伟达 H100 GPU 上,当求解一组千万阶矩阵的最大割半定规划松弛问题之时,只需要 10 秒到 1 分钟时间,而此前报道的 CPU 方法往往需要数十小时。

#cuLoRADS#Shanghai Jiao Tong University#PNAS#NVIDIA H100#Google Quantum AI#Nanjing University#Stanford University#MIT#MaxCut#NVIDIA

Original · 6 min readRead the original
Group chat

No comments yet. Start the conversation.