Stanford的Machine Learning又要开课了！

Stanford的Machine Learning又要开课了，有没有一起跟的小伙伴。


课程链接：https://www.coursera.org/course/ml

课程概述：

Machine learning is the science of getting computers to act without being explicitly programmed. In the past decade, machine learning has given us self-driving cars, practical speech recognition, effective web search, and a vastly improved understanding of the human genome. Machine learning is so pervasive today that you probably use it dozens of times a day without knowing it. Many researchers also think it is the best way to make progress towards human-level AI. In this class, you will learn about the most effective machine learning techniques, and gain practice implementing them and getting them to work for yourself. More importantly, you'll learn about not only the theoretical underpinnings of learning, but also gain the practical know-how needed to quickly and powerfully apply these techniques to new problems. Finally, you'll learn about some of Silicon Valley's best practices in innovation as it pertains to machine learning and AI.

This course provides a broad introduction to machine learning, datamining, and statistical pattern recognition. Topics include: (i) Supervised learning (parametric/non-parametric algorithms, support vector machines, kernels, neural networks). (ii) Unsupervised learning (clustering, dimensionality reduction, recommender systems, deep learning). (iii) Best practices in machine learning (bias/variance theory; innovation process in machine learning and AI). The course will also draw from numerous case studies and applications, so that you'll also learn how to apply learning algorithms to building smart robots (perception, control), text understanding (web search, anti-spam), computer vision, medical informatics, audio, database mining, and other areas.

课程视频：https://class.coursera.org/ml-005/lecture
课程schedule: https://class.coursera.org/ml-005/wiki/CourseSchedule

week1: http://www.1point3acres.com/bbs/thread-86319-1-1.html
week2: http://www.1point3acres.com/bbs/thread-87328-1-1.html
week3: http://www.1point3acres.com/bbs/thread-90861-1-1.html
week4: http://www.1point3acres.com/bbs/thread-92525-1-1.html
week5: http://www.1point3acres.com/bbs/thread-93793-1-1.html
week6: http://www.1point3acres.com/bbs/thread-95136-1-1.html
week7: http://www.1point3acres.com/bbs/thread-96002-1-1.html
week8: http://www.1point3acres.com/bbs/thread-96702-1-1.html
week9: http://www.1point3acres.com/bbs/thread-97270-1-1.html
week10: http://www.1point3acres.com/bbs/thread-97816-1-1.html

sanguine 发表于 2014-2-5 11:45
machine learning到底学些什么呢，能不能具体说说？我看看要不要学==

ML的范畴很大，现在做的人很多，以我现在的水平，不好概括啊。。。
“Machine learning is the science of getting computers to act without being explicitly programmed”，针对具体的问题不可能每一个都一一编程实现，ML就是想找到一个通用的方法。
比如：
1.让机器识别手写的数字。对人来说很容易，但是计算机不行，在计算机眼里所有的图像都是矩阵，要找到合适的算法来完成(比如神经网络)
2.对电子邮件分类，区分是正常邮件还是spam
3.推荐系统，比如：你在淘宝或者亚马逊买了东西，下次登录的时候它会推荐你可能喜欢的商品给你

ML里面很多模型是经典的统计学内容（比如：线性模型，logistic regression），对统计学出身的人来说偏应用偏实现，对于CS出身的人来说偏数学。事实上ML里面很多成果是CS的人提出来的，比如SVM。
有空就学学咯，这门课讲得很好理解，作业量不大（熟悉matlab的矩阵运算的话，每周的作业不超过3行代码。。）

sanguine 发表于 2014-2-5 11:45
machine learning到底学些什么呢，能不能具体说说？我看看要不要学==

Eroica已经说得很好了，我这个菜鸟就稍做补充哈。。。
就是用数据或以往的经验，来优化计算机程序的性能。其实就是各种经典的算法，决策树，SVM，Kmeans等等，主要分为监督学习和非监督学习，好像还有一个强化学习不知道是什么。
监督学习主要是regression和classification，比如你有很多组数据，分为输入值和输出值，从这组数据中学习出一个函数，这样有新的输入值就能预测结果了。举个例子，很简单single variable的线性回归，你有一组(x,y)，根据大量的(x,y)数据可以拟合出一个y=ax+b的函数，然后以后有新的x你就能得出y了。比如你根据以往的经验找到一个合适的spam/ham的邮件分类器，这样今后遇到新的邮件就可以区分了。
非监督学习就是你有输入值，但是输出值是unknown的，所以直接对输入值进行建模，根据数据内在的属性和联系进行划分，这部分主要是clustering，由于还没怎么学非监督学习，所以无法阐述太多哈。
如果走纯CS路线的话ML我觉得可以不用学，如果是想做数据分析那这个是必修课哦。

ML 是优化性能的吗？
不是机器根据已知数据预测未知的东西吗？
supervised是有部分数据含有要预测那个量的ground truth，来预测新数据里面那个未知量的值
比如上面说的linear regression, you know both x and y for some people, here y is the ground truth. you build a model to reflect the relationship between x and y.
and for the next group of people, you know only their x, and you want to know their y. then the model can do the prediction for you.  (linear model is used for more than prediction though).
the ground truth 通常是人工辨别的etc，


unsupervised就是没有这个ground truth，只看观察到的数据本身有什么特性

课程评价：http://coursetalk.org/coursera/machine-learning

sanguine 发表于 2014-4-14 06:27
我有的时候也是这样%>_

求加分，我已经把week 1 to 10都补上，除了几个penalty都是按时交的，麻烦了~

good news！！
上次跟了前一半，后一半就final了 没跟完。。

EroicaCMCS 发表于 2014-2-5 11:19
good news！！
上次跟了前一半，后一半就final了 没跟完。。

machine learning到底学些什么呢，能不能具体说说？我看看要不要学==

hahaa~太好了
上次跟了5周也没跟下去

上面写着3月10号开始啊？不是2月17啊

我了个大去，coursera不是说收藏了课程会有开课通知吗我怎么没收到！

跟了幾次都沒跟上，繼續跟啊！