简要声明
- 课程学习相关网址
- 由于课程学习内容为英文,文本会采用英文进行内容记录,采用中文进行简要解释。
- 本学习笔记单纯是为了能对学到的内容有更深入的理解,如果有错误的地方,恳请包容和指正。
- 非常感谢Andrew Ng吴恩达教授的无私奉献!!!
文章目录
专有名词
Machine Learning | 机器学习 | Supervised learning | 监督学习 |
---|---|---|---|
Unsupervised learning | 非监督学习 | Reinforcement learning | 强化学习 |
Regression | 回归问题 | Classification | 分类问题 |
label | 标签 | feature | 特征 |
cluster | 簇 | clustering algorithm | 聚类算法 |
Welcome to machine learning
Machine Learning
- Grew out of work in AI → 使机器自己学习如何去做有趣的事情
- New capability for computers
Examples
- Database mining 数据挖掘 (E.g., Web click data, medical records, biology, engineering)
- Applications can’t program by hand 无法手动编写程序 (E.g., Autonomous helicopter, handwriting recognition, Natural Language Processing (NLP), Computer Vision.)
- Self-customizing programs 私人订制程序 (E.g., Amazon, Netflix product recommendations)
- Understanding human learning 理解人类的学习过程
What is machine learning
Machine Learning definition
- Arthur Samuel (1959). Machine Learning: “Field of study that gives computers the ability to learn without being explicitly programmed.”
在没有明确设置的情况下,使计算机具有学习能力的研究领域。 - Tom Mitchell (1998) Well-posed Learning Problem: “A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E.”
字幕翻译:计算机程序从经验E中学习解决某一任务T进行某一性能度量P,通过P测定在T上的表现因经验E而提高。
DeepL:如果一个计算机程序在某个任务T和某个性能指标P方面的性能随着经验E的增加而提高,那么它就被称为从经验E中学习。
Machine learning algorithms
- Supervised learning & Unsupervised learning
- Reinforcement learning, recommender systems.
Supervised Learning
Housing price prediction
- 已知一个数据集,已知房子的平方英尺数,预测房子的价格。可以通过一条直线拟合数据,或者通过二次函数拟合数据。如何选择和如何决定用什么拟合数据会有更好的结果。
- Supervised Learning“right answers” given →数据集包含正确答案,算法基于正确答案预测更多的正确答案
- Regression: Predict continuous valued output (price) →回归问题:预测连续值输出
Breast cancer (malignant, benign)
- 已知一个数据集,已知肿瘤的大小,预测肿瘤是良性还是恶性的概率。
- Classification: Discrete valued output (0 or 1) →分类问题:预测一个离散值输出
- 机器学习会有很多feature特征
Unsupervised Learning
Unsupervised Learning
- 非监督学习没有任何labels标签
- 非监督学习会把数据分成不同的clusters簇
- 非监督学习是clustering algorithm聚类算法
Applications
- Organize computing clusters 组织大型计算机集群高效协作
- Social network analysis 社交网络分析
- Market segmentation 市场分割
- Segmentation data analysis 天文数据分析
吴恩达教授语录
- “Machine learning is one of the most exciting recent technologies.”
- “I like to make an analogy to learning to become a carpenter.”
- “I hope to make you one of the best people in knowing how to design and build serious machine learning and AI systems.”