毕 业 论 文题目 机器学习之线性回归与其正则化学院 理学院 专业 信息与计算科学年级 班级 黄剑华 学号 指导老师 职称 2024 年 6 月机器学习之线性回归与其正则化[摘要] 机器学习是计算机具备智能的基本途径，其最重要的核心为人工智能，对机器学习的运用遍布着人工智能的诸多领域，线性回归问题在监督机器学习的畴之。线性回归问题中，训练数据对未知数据的推测与模拟，即泛化能力可能不是很好，或者训练数据量不够，无法对训练数据集很好地回归分析，这时须要提高泛化能力，正则化便是提高泛化能力办法之一。正则化线性回归问题在于最小化误差的同时正则化参数，即模型拟合训练数据的同时，又须要制止模型过分地拟合训练数据，这样才具有良好的泛化性能。本文阐述了线性回归两种常见方法，梯度下降法与最小二乘法与其对应的正则化方法。通过数值模拟试验直观的理解线性回归与其正则化，与如何确定适合的正则化参数。[关键词] 线性回归模型 梯度下降法 最小二乘法 正则化线性回归 正则化参数Linear regression and regularization in machine learning[Abstract]Machine learning is the core of artificial intelligence . It is a fundamental way of intelligent computer , its applications across all areas of artificial intelligence , the linear regression problem belongs to the category of supervised machine learning . In linear regression problem , speculation and simulation the unknown data from training data , in other words , the generalization ability may not be very good , may also be the amount of training data is too small , not enough to regression analysis of the training data set , so it is necessary to improve the generalization ability of the model , regularization is a method to improve the generalization ability . Regularized linear regression problem is to minimize the error of simultaneous regularization parameters , minimize the error is to let the model fitting the training data and the regularization parameter is to prevent the model over fitting the ...
