熵正则推断：一种预测方法
摘要
预测推断需要在统计准确性与信息复杂性之间进行平衡，而复杂性度量通常被人为指定而非从理论中推导。我们将计量经济对象视为预测规则，即从信息映射到报告的预测分布的函数，并对评估提出三个结构性要求：局部性、严格良性以及在结果类别的聚合（细化/粗化）下的一致性。这些公理唯一地（最多经仿射变换）特征(logarithmic score)，并导致香农互信息（克洛德-戴维尔散度）成为对应的预测复杂性度量。 resulting entropy-regularized prediction problem admits Gibbs-form optimal rules, and we establish an essentially complete-class result for the admissible rules we study under joint risk-complexity dominance. Rational inattention emerges as the constrained dual, corresponding to frontier points with binding information capacity. The entropy penalty contributes additive curvature to the predictive criterion; in weakly identified settings, such as weak instruments in IV regression, where the unregularized objective is flat, this curvature stabilizes the predictive criterion. We derive a local quadratic (LAQ) expansion connecting entropy regularization to classical weak-identification diagnostics.
引用
@article{arxiv.2512.21637,
title = {Training-Free Disentangled Text-Guided Image Editing via Sparse Latent Constraints},
author = {Mutiara Shabrina and Nova Kurnia Putri and Jefri Satria Ferdiansyah and Sabita Khansa Dewi and Novanto Yudistira},
journal= {arXiv preprint arXiv:2512.21637},
year = {2025}
}