2015【Coursera】- Machine Learning by Andrew Ng （02-01 ~）
跟风开始Coursera的课程，其实去年就选过2门Coursera课程，但种种原因最终课程只听到一半就放弃了。
2015继续，选了专业相关的课程，而且和数学相关，曾经最喜欢的课程之一，希望能坚持下来。
Week 1 Due Sunday, February 8 at 23:59 PM PDT
- Introduction
- Linear Regression with One Variable
- (Optional) Linear Algebra Review
- Review Questions (for the week's topics)
20150201 ~ 20150209, ML的基础知识，听下来完全云里雾里，第一次测验2/5，中间还迷拼布了，到Review Question截止日课程还没听完，但意外第二次做Review Question却顺利通过，还在截止日前（截止日是美国时间），因此却激发了继续学习的兴趣动力，后面听下去倒是越听越有趣，还有Programming Exercise也是我喜欢的，喜欢边学便动手演练。
20150210, 借中午午餐时间把这节的课程听完，这节主要是线性代数在这门课中的应用，并温故基础线性代数的知识，蛮好懂的。大学时代学习线性代数都是用来做题、应付考试。这里能够把这门学科和实际的案例结合起来使用，把复杂的案例一下子变得简单易懂。
20150209 （4+4.25+5）
Week 2 Due Sunday, February 15 at 23:59 PM PDT
- Linear Regression with Multiple Variables 20150210
- Octave Tutorial 20150215
- Review Questions (for the week's topics)
- Programming Exercise 1 (Linear regression) 20150219
20150216 最近有些没跟上步调，在deadline前完成了review question，Programming最后赶在前天做完，扣掉了20%的惩罚分，还好题目都很简单，不过课程涉及线性代数、微积分等等，整体还没有完全的消化，也就依葫芦画瓢的过程，后面要拿出多点时间来内化，才能真正学懂这门课程。
Week 3 Due Sunday, February 22 at 23:59 PM PDT
- Logistic Regression 20150220
- Regularization
- Review Questions (for the week's topics)
- Programming Exercise 2 (Logistic regression)
20150222 赶在Deadline前完成了Review Question和Exercise，Week3的 Regularization 的课还没有听完, 只是看了一下PPT完成了Review Question.
课程听下来还是蛮感兴趣的，但专业知识比较多，需要课后花时间去理解，Programming Exercise越来越难做。看来额外阅读ML的课外参考书，把基础知识弄懂，否则后面的课程后举步维艰。
目前听英文看中文字幕听，如果没有中文字幕估计我要抓瞎了。
20150226 Regularization课程听完，后面找时间把Regularization的程序实现一遍，用来巩固基础知识的学习，Professor Andrew说学到这里如果课程都掌握了，那你就比很多学习ML的人懂得要多了，并且开始可以运用前面学到的知识来开发各种有趣的应用。
Week 4 Due Sunday, March 1 at 23:59 PM PDT
- Neural Networks: Representation
- Review Questions (for the week's topics)
- Programming Exercise 3
(Multi-class classification and neural networks)
20150226 Neural Networks: Representation 课程听完，神经网络的课程比前面的课程好玩，课上有趣的手写识别数字的程序很有趣，但复杂度增加了，而且基础知识不过关，这两周时间需要把前面的课程再加以巩固和消化，RQ和Exercise还没开始做，Exercise消耗我大量精力，计划留在周末做。
Week 5 Due Sunday, March 8 at 23:59 PM PDT
- Neural Networks: Learning
- Review Questions (for the week's topics)
- Programming Exercise
(Neural network learning)
Week 6 Due Sunday, March 15 at 23:59 PM PDT
- Advice for Applying Machine Learning
- Machine Learning System Design
- Review Questions (for the week's topics)
- Programming Exercise (Bias-variance)
Week 7 Due Sunday, March 22 at 23:59 PM PDT
- Support Vector Machines (SVMs)
- Review Questions (for the week's topics)
- Programming Exercise (SVMs)
Week 8 Due Sunday, March 29 at 23:59 PM PDT
- Clustering
- Dimensionality Reduction
- Review Questions (for the week's topics)
- Programming Exercise (K-Means and PCA)
Week 9 Due Sunday, April 5 at 23:59 PM PDT
- Anomaly Detection
- Recommender Systems
- Review Questions (for the week's topics)
- Programming Exercise
(Anomaly Detection and Recommender Systems)
Week 10 Due Sunday, April 12 at 23:59 PM PDT
- Large-Scale Machine Learning
- Example of an application of machine learning
- Review Questions (for the week's topics)
扩展阅读《Machine Learning in Action》
