← 研踪 PaperTrailarXiv:2605.07483 · v3
Does Your Neural Network Extrapolate? Feature Engineering as Identifiability Bias for OOD Generalization
首次提交 2026-05-08 17:25
版本更新 2026-11-07 03:53
雷达观测 2026-11-10 10:11
摘要
Deep neural networks can learn salient features from an in-distribution (ID) training window without identifying the data-generating structure needed for out-of-distribution (OOD) generalization. Understanding this limitation requires examining the identifiability of the data-generating process (DGP). From a single training window, the DGP that yields the correct OOD extrapolation is not identifiable: infinitely many DGPs are observationally equivalent within a fixed tolerance on that window yet diverge arbitrarily beyond it, and no ID criterion alone can reliably distinguish them. A structural commitment---the feature map, label map, and model class $(\varphi, ψ, \mathcal{M})$---specifies the assumed DGP and governs OOD generalization while leaving ID performance essentially unchanged. When architecture, pretraining, data augmentation, input formats, or domain knowledge implicitly supply the correct commitment, the model generalizes beyond the training window. When it cannot infer OOD-relevant structure from ID evidence, it fails to generalize. When the commitment is correct {\em and} identifiable, OOD error vanishes. We demonstrate this mechanism in models trained from scratch and in foundation models across four natural-science settings (chemistry, physics, biology, and histopathology), as well as in a positional-encoding study spanning Transformer, Mamba, and S4D architectures. Finally, a controlled study shows that correct features are necessary but not sufficient for OOD generalization.
期刊信息：The Fortieth Annual Conference on Neural Information Processing Systems 2026
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v3 · 2026-11-07 03:53 · 最新观测版本
Does Your Neural Network Extrapolate? Feature Engineering as Identifiability Bias for OOD Generalization
Deep neural networks can learn salient features from an in-distribution (ID) training window without identifying the data-generating structure needed for out-of-distribution (OOD) generalization. Understanding this limitation requires examining the identifiability of the data-generating process (DGP). From a single training window, the DGP that yields the correct OOD extrapolation is not identifiable: infinitely many DGPs are observationally equivalent within a fixed tolerance on that window yet diverge arbitrarily beyond it, and no ID criterion alone can reliably distinguish them. A structural commitment---the feature map, label map, and model class $(\varphi, ψ, \mathcal{M})$---specifies the assumed DGP and governs OOD generalization while leaving ID performance essentially unchanged. When architecture, pretraining, data augmentation, input formats, or domain knowledge implicitly supply the correct commitment, the model generalizes beyond the training window. When it cannot infer OOD-relevant structure from ID evidence, it fails to generalize. When the commitment is correct {\em and} identifiable, OOD error vanishes. We demonstrate this mechanism in models trained from scratch and in foundation models across four natural-science settings (chemistry, physics, biology, and histopathology), as well as in a positional-encoding study spanning Transformer, Mamba, and S4D architectures. Finally, a controlled study shows that correct features are necessary but not sufficient for OOD generalization.
查看此版本解读 →查看此版本原文 ↗