[论文] Soft Ontology Priors over Latent Concept Posteriors for Weakly Supervi...

论文概要

研究领域: ML
作者: Tengwei Song, Robert Hoehndorf
发布时间: 2026-10-24
arXiv: 2610.30391

中文摘要

只监督任务标签的神经符号（NeSy）预测器会学出「推理捷径」：不同的概念映射取得相同的最大似然，恢复出的概念不可辨识，下游复用不安全。本文向训练目标注入对潜概念的软本体先验，并把它与任务约束一起编译为光滑、可分解的概率电路。在独立伯努利感知下，所得结构概念后验的归一化子归约为加权模型计数，对编译后电路规模线性——因为神经叶子权重在求值前就折进了电路。软编译策略把专家公理放进电路的权重而非支撑：由数据生成过程保证的任务结构公理保持硬约束，噪声专家公义只能重分配先验质量、不可能清空支撑。文中刻画了先验如何通过本体违反能量之差重加权标签约束下的捷径质量，证明软先验在有界公理噪声下只产生有界的对数密度扰动，而硬支撑对应物可被单个损坏公理退化。在真实驾驶基准 BDD-OIA 的六概念审计视图与受控 NeSy 任务上验证了这些预测。

原文摘要

Neuro-symbolic (NeSy) predictors that supervise only the task label suffer from reasoning shortcuts: distinct concept mappings achieve the same maximum likelihood, leaving the recovered concepts unidentifiable and unsafe for downstream reuse. We attack this failure by injecting a soft ontology prior over the latent concepts into the training objective and compiling it together with the task constraint into a smooth, decomposable probabilistic circuit. Under independent Bernoulli perception, the normalisers of the resulting structured concept posterior reduce to weighted model counts that are linear in the compiled circuit size, because the neural leaf weights fold into the circuit before evaluation. The soft compilation strategy places expert axioms in the circuit's weights rather than its support (task-structural axioms guaranteed by the data-generating process stay hard), so noisy expert axioms only redistribute prior mass and cannot empty the support. We characterise how the prior reweights label-restricted shortcut mass through differences in ontology violation energy, show that soft priors incur bounded log-density perturbations under bounded axiom noise, and prove that the hard-support counterpart can be made degenerate by a single corrupted axiom. We validate these predictions on an audited six-concept view of the real-world BDD-OIA driving benchmark and on controlled NeSy tasks.

自动采集于 2026-10-24

#论文 #arXiv #ML #小凯