用多项式混沌代理模型分解生成模型中不确定性的来源，让决策依据可解释。
Interpretable epistemic uncertainty decomposition in sequential generative models via polynomial chaos surrogates
- 通过多项式混沌展开拟合小规模模型集，解析奖励不确定性对生成过程的影响。
- 在催化剂选择任务中，配体稳定性指数达179，是配体的2.5倍，揭示脆弱环节。
- 生成速度比重训练快1000~10000倍，且结果可通过形式化验证，适合高可靠性场景。
基于不确定奖励的序列生成模型在人工智能驱动的科学发现中至关重要，但其继承的表征不确定性尚未被量化。本文通过将多项式混沌展开（PCE）拟合到少量训练好的模型上，将此类不确定性传播至生成流网络（GFlowNets）。PCE系数可解析出各奖励分量对生成决策的敏感度，首次实现可解释的不确定性分解，这是深度集成、贝叶斯神经网络或蒙特卡洛丢弃无法做到的。理论证明了收敛性，其中四项已在Lean 4形式化证明系统中严格验证。在三个真实任务中，该框架揭示了集成方法无法发现的可操作结构：在Doyle-Dreher Buchwald-Hartwig数据集上，催化剂选择稳健（$D_{\mathrm{catalyst}}\approx 71$），而添加剂选择脆弱（$D_{\mathrm{additive}}\approx 179$，为前者的2.5倍）；在片段分子设计中，连接子位置最敏感（$D_{\mathrm{linker}}\approx 28$），修饰位点最稳健（$D\approx 14$-$18$），推翻传统“骨架稳健/修饰脆弱”假设；在Sachs蛋白信号网络中，MAPK通路边与PKA/PKC枢纽边呈现不同敏感度，可指导靶向扰动实验。95%置信水平下校准覆盖率达0.97–1.00，且代理模型可在毫秒级评估10,000条策略样本，较完全重训练快10³–10⁴倍。
原文摘要 · Abstract (English)
Sequential generative models conditioned on uncertain rewards are central to AI-driven scientific discovery, yet the epistemic uncertainty they inherit from imperfect reward estimates remains unquantified. We propagate this uncertainty through generative flow networks (GFlowNets) by fitting polynomial chaos expansions (PCEs) to small ensembles of trained models. The PCE coefficients yield analytical Sobol sensitivity indices, providing the first interpretable decomposition of which reward components drive which generative decisions, a capability unavailable from deep ensembles, Bayesian neural networks, or Monte Carlo dropout. Convergence guarantees are established theoretically and four of five are formally verified in the Lean 4 proof assistant. Across three real-world tasks the framework reveals actionable structure invisible to ensembles alone. On the Doyle-Dreher Buchwald-Hartwig dataset catalyst selection is robust ($D_{\mathrm{catalyst}}\approx 71$) while additive selection is fragile ($D_{\mathrm{additive}}\approx 179$, $2.5\times$ higher). In fragment-based molecular design the linker position is the most sensitive ($D_{\mathrm{linker}}\approx 28$) while decoration positions are the most robust ($D\approx 14$-$18$), reversing the conventional scaffold-robust / decoration-fragile assumption. On the Sachs protein signalling network, MAPK-cascade edges and PKA/PKC hub edges separate into distinct sensitivity regimes, providing a targeted map for perturbation experiments. Calibration coverage at the 95% level reaches 0.97-1.00 across the dominant steps, and the surrogate evaluates 10{,}000 policy samples in milliseconds - $10^{3}$-$10^{4}\times$ faster than exhaustive retraining.
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。