\noindent Hyper-parameter selection is a central practical problem in modern machine learning, governing regularization strength, model capacity, and robustness choices. Cross-validation is often computationally prohibitive at scale, while fully Bayesian hyper-parameter learning can be difficult due to the cost of posterior sampling. We develop a generative perspective on hyper-parameter tuning that combines two ideas: (i) optimization-based approximations to Bayesian posteriors via randomized, weighted objectives (weighted Bayesian bootstrap), and (ii) amortization of repeated optimization across many hyper-parameter settings by learning a transport map from hyper-parameters (including random weights) to the corresponding optimizer. This yields a ``generator look-up table'' for estimators, enabling rapid evaluation over grids or continuous ranges of hyper-parameters and supporting both predictive tuning objectives and approximate Bayesian uncertainty quantification. We connect this viewpoint to weighted $M$-estimation, envelope/auxiliary-variable representations that reduce non-quadratic losses to weighted least squares, and recent generative samplers for weighted $M$-estimators.
翻译:超参数选择是现代机器学习中的核心实际问题,它决定了正则化强度、模型容量与鲁棒性选择。在大规模场景下,交叉验证的计算成本往往过高,而完全贝叶斯超参数学习则因后验采样代价而难以实现。本文提出一种超参数调优的生成式视角,该视角融合了两个核心理念:(i) 通过随机加权目标(加权贝叶斯自助法)实现基于优化的贝叶斯后验近似;(ii) 通过学习从超参数(包括随机权重)到对应优化器的传输映射,对大量超参数配置中的重复优化过程进行摊销。该方法构建了估计量的“生成器查找表”,支持在网格或连续超参数范围内快速评估,既能满足预测性调优目标,又能实现近似贝叶斯不确定性量化。我们将该视角与加权$M$估计、将非二次损失转化为加权最小二乘的包络/辅助变量表示,以及近期针对加权$M$估计量的生成式采样器建立了理论联系。