T-范数运算符用于欧盟人工智能法案合规分类:在神经符号推理系统中对Lukasiewicz、Product和Gödel语义的实证比较
T-Norm Operators for EU AI Act Compliance Classification: An Empirical Comparison of Lukasiewicz, Product, and Gödel Semantics in a Neuro-Symbolic Reasoning System
AI总结:
研究比较了三种T-范数运算符在神经符号推理系统中的表现,发现Gödel运算符在准确性与召回率上表现最佳,但存在少量误报,而Product和Lukasiewicz运算符在零误报情况下表现较弱。
摘要:
我们展示了三种t-范数运算符——Lukasiewicz(T_L)、Product(T_P)和Gödel(T_G)——作为逻辑合取机制在欧盟人工智能法案合规分类神经符号推理系统中的首次比较性试点研究。使用LGGT+(逻辑引导图变换器Plus)引擎和包含四个风险类别(禁止、高风险、有限风险、最低风险)的1035个注释AI系统描述基准,我们评估了分类准确率、误报率和误检率以及运算符在模糊情况下的行为。在n=1035时,三种运算符显著不同(McNemar p<0.999)。T_G实现最高准确率(84.5%)和最佳边界召回率(85%),但引入8个误报(0.8%)通过min-semantics过分类。T_L和T_P保持零误报,其中T_P优于T_L(81.2% vs. 78.5%)。我们的主要发现是:(1)运算符选择次于规则库完整性;(2)T_L和T_P保持零误报但遗漏边界情况;(3)T_G的min-semantics在召回率上更高,但以0.8%的误报率为代价;(4)混合semantics分类器是下一步的可行方向。我们释放LGGT+核心引擎(201/201测试通过)和基准数据集(n=1035)在Apache 2.0协议下。
英文摘要:
We present a first comparative pilot study of three t-norm operators -- Lukasiewicz (T_L), Product (T_P), and Gödel (T_G) - as logical conjunction mechanisms in a neuro-symbolic reasoning system for EU AI Act compliance classification. Using the LGGT+ (Logic-Guided Graph Transformers Plus) engine and a benchmark of 1035 annotated AI system descriptions spanning four risk categories (prohibited, high_risk, limited_risk, minimal_risk), we evaluate classification accuracy, false positive and false negative rates, and operator behaviour on ambiguous cases. At n=1035, all three operators differ significantly (McNemar p<0.001). T_G achieves highest accuracy (84.5%) and best borderline recall (85%), but introduces 8 false positives (0.8%) via min-semantics over-classification. T_L and T_P maintain zero false positives, with T_P outperforming T_L (81.2% vs. 78.5%). Our principal findings are: (1) operator choice is secondary to rule base completeness; (2) T_L and T_P maintain zero false positives but miss borderline cases; (3) T_G's min-semantics achieves higher recall at cost of 0.8% false positive rate; (4) a mixed-semantics classifier is the productive next step. We release the LGGT+ core engine (201/201 tests passing) and benchmark dataset (n=1035) under Apache 2.0.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。