Scholarlink

The introduction of regularization techniques and neural networks in the field of Machine LearningMachine learning has garnered substantial research. This provides sufficient knowledge on the effects of dropoutsDropout and optimizersOptimizer in the case of classical neural networks. However, the consequences of the same using logistic regressionLogistic regression, across various interaction degreesInteraction degree, haven’t been thoroughly studied. In this paper, we aim to discuss the effects of dropoutsDropout with different optimizersOptimizer (of different values for the hyperparameters—momentumMomentum and dropoutDropout probability) on interaction degreesInteraction degree varying from 1 to 3 and demonstrate, both mathematically and empirically, that the best performance for a logistic regressionLogistic regression model is a combination of Adam optimizerOptimizer and Gaussian dropoutsDropout with Interaction degreeInteraction degree = 2.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Interaction Modeling in Logistic Regression: An Analysis of Dropout Regularization and Adaptive Optimization

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Niveditha Kundpuram,
  • Subha Vidisha Atilli,
  • K. S. Srinivas

摘要

The introduction of regularization techniques and neural networks in the field of Machine LearningMachine learning has garnered substantial research. This provides sufficient knowledge on the effects of dropoutsDropout and optimizersOptimizer in the case of classical neural networks. However, the consequences of the same using logistic regressionLogistic regression, across various interaction degreesInteraction degree, haven’t been thoroughly studied. In this paper, we aim to discuss the effects of dropoutsDropout with different optimizersOptimizer (of different values for the hyperparameters—momentumMomentum and dropoutDropout probability) on interaction degreesInteraction degree varying from 1 to 3 and demonstrate, both mathematically and empirically, that the best performance for a logistic regressionLogistic regression model is a combination of Adam optimizerOptimizer and Gaussian dropoutsDropout with Interaction degreeInteraction degree = 2.