ProCerDCL：面向深度学习编译器优化的查询高效黑盒概率验证框架
ProCerDCL: Query-Efficient Black-Box Probabilistic Verification of Deep Learning Compiler Optimization
AI 导读
arXiv 论文提出 ProCerDCL，一个查询高效的黑盒概率验证框架，在给定输入分布和二元属性 oracle 下，以三值判定（接受 p≤θ、拒绝 p>θ+η、不确定）评估深度学习编译器优化的属性违反概率，各类错误判定概率由设定的误差预算约束。
正文 · 原文
Abstract:We present \tool, a query-efficient framework for black-box probabilistic verification. Given a fixed, sampleable input distribution and a binary property oracle, \tool assesses the property's violation probability $p$ against a threshold $\theta$. Under independent sampling and stateless oracle behavior, acceptance asserts $p\le\theta$, rejection asserts $p>\theta+\eta$ for a user-specified separation margin, and an inconclusive verdict indicates insufficient evidence for either conclusion. The probability of each type of erroneous verdict is bounded by its specified error budget. To support this three-valued contract efficiently, \tool uses a single sequential test, eliminating the interval family required by the baseline. The test accounts for repeated sampling checks through either corrected exact binomial confidence intervals or an anytime-valid confidence sequence. We apply \tool to probabilistic local model semantic consistency, where the property oracle detects decision-level inconsistencies between reference and compiled executions under a specified perturbation distribution. Our evaluation covers 24 tasks across TorchInductor and TensorRT and examines sampling efficiency, conclusiveness, and diagnostic behavior. For sampling efficiency, we use paired one-model clean-label radius searches: the baseline requires 3.6-4.2 times as many unique samples under matched numerical parameters while providing a weaker decision contract. For model semantic consistency, checks at $\theta=10^{-3}$ yield 817 acceptances, 19 inconclusive verdicts, and 4 rejections across 840 input--backend instances from 21 tasks. Case studies of two documented compiler faults further demonstrate what these local verdicts can diagnose.
| Subjects: | Software Engineering (cs.SE); Artificial Intelligence (cs.AI); Programming Languages (cs.PL) |
| Cite as: | arXiv:2610.36499 [cs.SE] |
| (or arXiv:2610.36499v1 [cs.SE] for this version) | |
| https://doi.org/10.48550/arXiv.2610.36499 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Simin Chen [view email]
[v1]
Tue, 27 Oct 2026 01:57:08 UTC (840 KB)
来源：arXiv cs.AI · arxiv.org