论文标题
Fisher标准的扩展：具有神经网络实现的理论结果
An Extension of Fisher's Criterion: Theoretical Results with a Neural Network Realization
论文作者
论文摘要
Fisher的标准是用于特征选择的机器学习中广泛使用的工具。对于大型搜索空间，Fisher的标准可以为选择功能提供可扩展的解决方案。然而，对Fisher标准的一个挑战性局限性是，当班级条件分布的平均值彼此接近时，它的性能差。在这一挑战的推动下，我们提出了Fisher标准的扩展，以克服这一限制。提出的扩展利用了类条件分布的可用异质性，将一个类与另一个类区分开。此外，我们描述了如何将我们的理论结果施加到神经网络框架中，并进行概念验证实验，以证明我们解决分类问题方法的生存能力。
Fisher's criterion is a widely used tool in machine learning for feature selection. For large search spaces, Fisher's criterion can provide a scalable solution to select features. A challenging limitation of Fisher's criterion, however, is that it performs poorly when mean values of class-conditional distributions are close to each other. Motivated by this challenge, we propose an extension of Fisher's criterion to overcome this limitation. The proposed extension utilizes the available heteroscedasticity of class-conditional distributions to distinguish one class from another. Additionally, we describe how our theoretical results can be casted into a neural network framework, and conduct a proof-of-concept experiment to demonstrate the viability of our approach to solve classification problems.
