发表机构
Indian Institute of Technology Guwahati; SAFIR, Sorbonne University Abu Dhabi; Indian Institute of Technology Bombay（印度技术学院加瓦哈蒂分校; 阿布扎比索邦大学 SAFIR; 印度技术学院孟买分校）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出用于表格回归的LAIR-Net，通过泄漏残差变换混入浅层锚点，在23个基准数据集上的平均排名优于8个随机网络和12个传统模型，性能与非线性结构及噪声量相关。
AI 中文摘要
深度随机模型通过随机初始化固定隐藏层参数，仅学习闭式读出，通常通过堆叠随机变换增加深度，却未对隐藏状态演化进行目标感知控制。我们提出LAIR-Net，即泄漏对齐脉冲残差网络，其通过泄漏残差变换将浅层学习的锚点混入每个隐藏状态。我们推导了输入扰动敏感性的深度一致界，并通过受控模拟将其相对于随机基线的性能提升归因于锚点，而非递归或增加的容量。当在可用噪声水平下可学习非线性目标结构时，该方法的优势显现；对于近线性目标或主导噪声，优势则减弱。在23个基准数据集上，LAIR-Net在8个随机网络和12个传统模型中取得最佳平均排名，其相对性能与模拟中识别的相同非线性结构和噪声量相关。
英文摘要
Deep randomized models fix hidden-layer parameters through random initialization and learn only closed-form readouts, typically adding depth by stacking random trans formations without target-aware control of hidden-state evolution. We propose LAIR Net, the Leaky Alignment-Impulse Residual Network, which mixes a shallow learned anchor into each hidden state through a leaky residual transition. We derive a depth uniform bound on input-perturbation sensitivity and use controlled simulations to attribute gains over a randomized baseline to the anchor rather than recursion or added capacity. Benefits emerge when a nonlinear target structure is learnable at the available noise level and diminish for nearly linear targets or dominant noise. Across 23 benchmark datasets, LAIR-Net achieves the best average rank among eight randomized networks and twelve conventional models, with relative performance associated with the same nonlinear-structure and noise quantities identified in simulation.
