用敲击法控制深度网络变量筛选的假阳性，提升模型简洁性
Knockoffs-based False Discovery Rate Control and Simplification for Deep Neural Networks
- 基于敲击法与正则化网络设计三类变量筛选方法
- 在控制假发现率前提下有效识别无关参数与输入变量
- 适合需要可解释性与计算效率的深度学习应用
深度神经网络在机器学习中广泛应用，但常包含大量参数和输入变量，其中许多与目标输出无关。这些无关变量不仅增加计算复杂度，还带来额外计算开销。为解决此问题，本文借鉴敲击法（knockoff methods）并结合正则化神经网络，提出三种在控制假发现率条件下的变量筛选方法：单层过滤、多层过滤和变量权重聚合过滤。实验表明，相较于现有算法，所提方法在保持低假阳性率的同时，具备良好筛选性能。
原文摘要 · Abstract (English)
The deep neural network is a widely used framework in machine learning that has been widely applied in various fields. However, deep neural networks often involve a large number of parameters and inputs, many of which may be irrelevant to the goal or true output. These parameters and input variables not only increase computational complexity, but also contribute to additional computational cost. One solution to this problem is knockoff methods, which have proven successful in controlling false discovery rates in high-dimensional regression. Building on the knockoff methods and using the regularised neural network, this paper proposes three variable screening methods under the condition of controlling false discovery rates: one layer filter, multiple layers filter, and variable weight aggregation filter. In comparison with existing algorithms, we find that our algorithms show satisfactory performance.
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。