用高斯分布优化调度，让编译器更高效地自动排程。
GauS: Differentiable Scheduling Optimization via Gaussian Reparameterization
- 将调度问题转为连续高斯变量，捕捉时间顺序关系。
- 在多个基准上实现帕累托最优，优化空间大幅缩小。
- 适合需要自动调度的编译器与硬件合成研究者。
高效的操作符调度是软件编译与硬件综合中的核心挑战。尽管近期可微方法试图用基于梯度的搜索替代传统方法（如精确求解器或启发式算法），但通常依赖类别分布，无法体现时间的有序性，且参数空间扩展性差。本文提出一种新颖的可微框架 GauS，通过高斯分布对操作符调度进行随机松弛，充分利用现代并行计算设备（如 GPU）。通过将调度表示为连续的高斯变量，成功捕捉了时间的有序特性，并使优化空间缩小数个数量级。该方法高度灵活，可表达多种目标与约束，首次为复杂的流水线调度问题提供了可微形式化。我们在多个基准上评估了该方法，结果表明 GauS 能达到帕累托最优性能。
原文摘要 · Abstract (English)
Efficient operator scheduling is a fundamental challenge in software compilation and hardware synthesis. While recent differentiable approaches have sought to replace traditional ones like exact solvers or heuristics with gradient-based search, they typically rely on categorical distributions that fail to capture the ordinal nature of time and suffer from a parameter space that scales poorly. In this paper, we propose a novel differentiable framework, GauS, that models operator scheduling as a stochastic relaxation using Gaussian distributions, which fully utilize modern parallel computing devices like GPUs. By representing schedules as continuous Gaussian variables, we successfully capture the ordinal nature of time and reduce the optimization space by orders of magnitude. Our method is highly flexible to represent various objectives and constraints, which provides the first differentiable formulation for the complex pipelined scheduling problem. We evaluate our method on a range of benchmarks, demonstrating that Gaus achieves Pareto-optimal results.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。