|
所在平台: Coursera |
课程主页: https://www.coursera.org/learn/ntumlone-algorithmicfoundations
课程评论:没有评论
课程名称:机器学习基石下(Machine Learning Foundations)——算法基础 概述:机器学习是研究如何让计算机通过从观察到的数据中积累经验而自我改进的一个领域。我们的两门姐妹课程介绍了机器学习用户应掌握的基础算法、理论和实践工具。本课程第二部分将更侧重于算法工具,而另一门课程则更侧重于数学工具。 课程大纲: - 第九讲:线性回归 描述:通过解析解即时计算线性假设的权重向量和平方误差。 - 第十讲:逻辑回归 描述:在交叉熵误差上进行梯度下降,以获得良好的逻辑假设。 - 第十一讲:分类的线性模型 描述:通过(逻辑)回归进行二分类;通过一对多(OVA)/一对一(OVO)分解进行多分类。 - 第十二讲:非线性转换 描述:通过非线性特征变换 + 线性模型实现非线性模型,同时处理模型复杂性的代价。 - 第十三讲:过拟合的危害 描述:当模型过于复杂、存在随机/确定性噪声和数据有限时,会发生过拟合。 - 第十四讲:正则化 描述:最小化增强误差,其中添加的正则项有效地限制了模型复杂性。 - 第十五讲:验证 描述:使用(交叉)保留验证数据来模拟模型选择的测试过程。 - 第十六讲:三项学习原则 描述:关注模型复杂性、数据质量以及你的专业水平。
Name: 第九講: Linear Regression
Description:weight vector for linear hypotheses and squared error instantly calculated by analytic solution
Name:第十講: Logistic Regression
Description:gradient descent on cross-entropy error to get good logistic hypothesis
Name:第十一講: Linear Models for Classification
Description:binary classification via (logistic) regression; multiclass classification via OVA/OVO decomposition
Name:第十二講: Nonlinear Transformation
Description:nonlinear model via nonlinear feature transform+linear model with price of model complexity
Name:第十三講: Hazard of Overfitting
Description:overfitting happens with excessive power, stochastic/deterministic noise and limited data
Name:第十四講: Regularization
Description:minimize augmented error, where the added regularizer effectively limits model complexity
Name:第十五講: Validation
Description:(crossly) reserve validation data to simulate testing procedure for model selection
Name:第十六講: Three Learning Principles
Description:be aware of model complexity, data goodness and your professionalism
Machine learning is the study that allows computers to adaptively improve their performance with experience accumulated from the data observed. Our two sister courses teach the most fundamental algorithmic, theoretical and practical tools that any user of machine learning needs to know. This second course of the two would focus more on algorithmic tools, and the other course would focus more on mathematical tools. [機器學習旨在讓電腦能由資料中累積的經驗來自我進步。我們的兩項姊妹課程將介紹各領域中的機器學習使用者都應該知道的基礎演算法、理論及實務工具。本課程將較為著重方法類的工具,而另一課程將較為著重數學類的工具。]