Save for later

Machine Learning

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.

Get Details and Enroll Now

OpenCourser is an affiliate partner of Coursera.

Get a Reminder

Not ready to enroll yet? We'll send you an email reminder for this course

Send to:



Stanford University

Rating 4.9 based on 28,977 ratings
Length 12 weeks
Starts May 11 (3 weeks ago)
Cost $79
From Stanford University via Coursera
Instructor Andrew Ng
Download Videos On all desktop and mobile devices
Language English
Subjects Programming Data Science
Tags Computer Science Data Analysis & Statistics Data Science Machine Learning

Get a Reminder

Get an email reminder about this course

Send to:

Similar Courses

What people are saying

According to other learners, here's what you need to know

machine learning in 8609 reviews

This course is a good first step of the specialization in machine learning.

Each week one lecturer explains the "idea" behind a machine learning algorithm, then the other one implements parts of it.

Although Machine learning has run several times since its first offering and it doesn’t seem to have been changed or updated much since then, it holds up quite well.

I was initially turned off by the use of MATLAB/Octave as the programming language of choice for the assignments, but I found them relatively painless and well-crafted to give the student a modular view of how these machine learning algorithms work and the possible optimizations when implementing them.

Because I was already familiar with most of the methods in the beginning (linear and multiple regression, logistic regression), I could focus more on the machine learning perspective that the class brought to these methods.

It is a very good course for anyone who wants to begin their journey into Machine Learning.

A fairly good overview of machine learning, with a fair amount of breadth but almost no depth.

All other Machine Learning courses require an advanced knowledge of programming, this one is not, and I really appreciate it as I have a background in statistics but not much coding experience .

Great introduction to machine learning.

This is a watered-down course of Machine Learning.

However, if you are a beginner, this course is a great way to start learning about Machine Learning.

I published my thoughts on the course and its contents on in a blog post which you can find here: A great introduction to machine learning.

Machine learning course was the best courses I ever found.

Andrew is a great teacher on Machine Learning, and he presents just the right amount of math that you need to know in order to understand the lecture.

Read more

so much in 671 reviews

Great Class I love this course from here i learn so much about machine learning and the implementation on our life Great introduction, clear explanations.

很赞,老师讲解的很细致,作业很有针对性 Thank you so much 吴恩达老师, you took me into Machine Learning world.

Thank you so much for offering it for free, and for not being so strict with the deadlines.

Thank you Coursera and Andrew NG for compiling so much knowledge in 1 course.

Thank you so much coursera..!!

Mr. Andrew Ng and other mentors helped so much to understand this course.

I like it so much, that now, I will have to take all the next ones (and I am already sure it will be a pleasure).

Very nice course, I learned so much!

I learnt so much about machine learning principles in general (and able to work on some machine learning problems).

one of the best courses on ML across the web Very good machine learning course, I learned so much from this course.

Thanks you so much for putting together this class.

I am just loving this course so much.

Thank you so much Prof. Andrew Ng.

really helpful course, I enjoyed it so much, it's easy to watch and understand.

Read more

decision trees in 32 reviews

讲的通俗易懂,理论与实践相结合,非常适合机器学习刚入门的小白,强烈推荐! Lack of decision trees but overall course is super-duper!

It would be great if decision trees topic is also included in the course.

It just misses decision trees I thank the instructor Andrew Ng Sir and Coursera for hosting such a wonderful course.

Missing modules for Decision Trees, Naive Bayes and more about Deep Learning.

I would have appreciated a dive into decision trees or boosted decision trees, however.

Content was OK, but quality of teaching was fair at best -- important points glossed over, many not made clear at all, some simply omitted: Bayes classifiers, decision trees, etc, etc..

Some other fine machine learning algorithms could be presented here (decision trees, bayesian networks, hidden markov chains, genetics algorithms, etc), nevertheless it is understandable that the course has a specific scope either in a group of topics either in time, so no regrets about not finding more machine learning subjects here, it represents my new learning backlog for the future.

Decision trees, and some of the models recently in vogue (such as CNN) aren't covered at all.

However, I personally feel that "DECISION TREES" is the only frequently used technique in ML that is missing in this course.

So, adding Decision trees Concept too would be very helpful and makes the course an equivalent master level course.

Some popular techniques such as decision trees and ensemble learning are not touched at all.It's a very well thought out course.

Machine learning techniques like decision trees and ensemble methods which are not covered here.

I really enjoyed the coding parts: guided, thorough and informative, with a lot of optional exercices to test one's comprehension of the material.I'd have liked to hear about decision trees, but hey: that would have been the cherry ON the cherry on the cake.Thank you Professor Ng for making this course available free to all learners around the world.

The only regret I feel that the course does not cover decision trees and bayesian networks.

Read more

random forest in 30 reviews

The willingness to dive into linear algebra is an recommended prequisite.Course gives not only insight into different kind of Machine Learning model paradigma but also offers practival advice for implementation of ML systems.Course does not coffer decision tree or random forest etc classifiers, which might have been nice.All in all, I can wholeheartedly recommend taking this course.

Please include Tree Based Models like Decision Trees, Random Forests etc.

If adding more machine learning algorithms such as decision trees/random forests should be better.

Awesome couse, I feel Decssion trees and random forest algos also should be included.

And even though this course does not touch on all of the significant ML methods (e.g., random forests), it definitely delves (a purposefully chosen verb, mind you) into perhaps the most significant.

Also, important/popular algorithms like Random Forest were not included.

I'd like to see more algorithms covered - naive bayes, decision trees, random forests, more examples on neural networks (RNN, CNN) Very help.

Please add the random forest algorithm.

Absolutely the best introductory course to machine learning!Wish the course also covered decision trees, random forests and boosting algorithms.

Excellent course as it is, but still would appreciate the inclusion of topics such as Monte Carlo Tree Search and Random Forest search as those also seem to be used often enough in ML settings.

using caret, glm, random forest etc in R. One of the most impressive courses I ever had.

I also a little surprised that Random Forests and Decision tree based algos were not covered in the course.

Needs to include decision trees - random forests and xgboost - which have become popular since this course was produced.

tree based algorithms, random forest etc.

Read more

recommender systems in 25 reviews

(ii) Unsupervised learning (clustering, dimensionality reduction, recommender systems, deep learning).

I only wish there we more programming exercises for the areas on Neural networks, SVM and recommender systems.

digits recognition, compression algorithms, and recommender systems).

By learning how to actually write the code, I now feel I understand to a much deeper level how facial recognition, recommender systems and autonomous driving works.

I had difficulties with the programming exercises for forward and backward propagation as well as with recommender systems.

But for someone who can handle all the notations and math (mostly linear algebra and optimization algorithms), this course is an ideal introduction into the concepts of Machine Learning (Supervised, Unsupervised & Recommender Systems, as well as diagnostic tools).

I really wanted that certificate and the quiz on Recommender Systems/Collaborative filtering should really be re-worked.

Barring a few topics in the end such as recommender systems which I feel was a topic too complex but had to be brushed through, I genuinely enjoyed every other one.

Now, for example, when visiting web sites I view recommender systems with a much better understanding of what they are doing behind the scenes.

It covers both Supervised and Unsupervised learning as well as some more advanced topics (Recommender Systems, Large Scale Machine Learning).

Same thing about recommender systems and spam-filtering.

Inspiring me to pursue Introduction to Recommender Systems at Coursera.

Unsupervised learning (clustering, dimensionality reduction, recommender systems, deep learning).

The examples in lectures and programming assignments are from real world applications and provides a great feeling to know how some of the things I use actually work behind the scenes, like recommender systems for movies/products, image to text conversion systems, image compresionm market segmentation etc.

Read more

anomaly detection in 21 reviews

I strongly believe this course is a must have if you're starting in neural network, but also great for any kind of machine learning problem like image recognition, data classifier and anomaly detection, function modeling and so on.

Neural Networks or Anomaly Detection).

K means and anomaly detection.Anomaly detection is explained very well plus a example is given and solved whereas in K-means he said that we will assign points to clusters and move clusters but he didn't perform calculations so it was difficult for me to do calculations for this topic.

Especially in the Anomaly Detection, the usage of PDF was very ambiguous.

I also learned what I've learned when I was in the university like anomaly detection and neural network.

Topics like supervised/unsupervised learning, Anomaly Detection, Computer Vision though harsh are intuitively introduced to learners and clearly explained.

This course is perfect,I enrolled in April, and finished in August.In the beginning, I don't think this course will help me a lot in machine learning, but now I think what I thought in the beginning is absolutely wrong, I learned very much in this course, including logistic regression, linear regression, regularization, bias and variance, anomaly detection, principal component analysis, clustering algorithm-keans, recommender system, multiclassification, support vector machine, kernel methods, gaussian distribution, stochastic gradient descent, map-reduce, artificial neural network and backpropagation octave programming etc.If you want to break into machine learning, I think this course will help you to do so.

fraud detection) via anomaly detection algorithms.

You will learn about and apply many concrete machine learning algorithms (neural networks, linear and logistic regression, anomaly detection and others).

computer vision and anomaly detection.

I wish I had more teacher like this when I was a student.I wish there would be a follow up with maybe even more tips for large scale machine learning and datasets where the positive class is only a small fraction of all the data available (not anomaly detection though).Thanks a lot Andrew !

Now off to do some simple applications here at work like spam filter and anomaly detection to start.

机器学习入门课程,Andrew对许多算法的原理做了直观性的介绍,布置的作业也对深入理解很好的促进;motivation和application的介绍非常棒,跟时代结合紧密,比如spam和anomaly detection;内层的数学没有较多的介绍,所以整个课程上下来不会很疲惫,时间长度比较合理;建议老师在结课时推荐一些更深入学习机器学习的进阶课程。最后非常感谢老师的教学,这门课让人受益良多 很好的入门课程,调理清晰 It really is a great class!

Typical algorithms such as supervised and unsupervised learning algorithms, support vector machine, PCS, Clustering, Anomaly Detection, and so on were taught and explained completely.

Read more

support vector machines in 15 reviews

Topics include: (i) Supervised learning (parametric/non-parametric algorithms, support vector machines, kernels, neural networks).

He takes you through behind-the-scenes of magic of regression, classification, clustering, neural networks, support vector machines in a very lucid manner.

Enjoyable, easy-to follow and yet quite detailed introduction to things like logistic regression, neural networks, support vector machines, or recommender systems.

Turns out machine learning is just math, and this course teaches the calculus & linear algebra concepts needed to understand linear regression, neural networks, support vector machines, K-means clustering, and more.

In that regard, I found the lectures on support vector machines sadly very confusing (I learned more by downloading Andrew Ng's lectures notes from his actual Stanford course).It's nevertheless a good introductory course and I would recommend it to anybody who wants to learn the basics of machine learning.

For example, for Support Vector Machines and backpropagation for neural networks.

Andrew was a very good teacher, taking time to explain fundamentals of linear regression, logistic regression, neural networks, support vector machines, case studies and more.

However, the treatment of SVM (Support Vector Machines) was a bit too light for my tastes.

Mind you, I'm biased because I was a math major and want to see proofs for everything, but I would have really liked to see more of the details behind support vector machines and neural networks.

This course is an opportunity to get acquainted with several machine learning techniques, including linear regression, logistic regression, Support Vector Machines (SVM), anomaly detection, non-supervised learning (clustering, K-means, etc), recommendation systems and very interesting discussions about batch/mini-batch versus stochastic learning and large-scale learning systems.

Topics include: Supervised learning (parametric/non-parametric algorithms, support vector machines, kernels, neural networks).

The course runs 10 weeks and covers a variety of topics and algorithms in machine learning including gradient descent, linear and logistic regression, neural networks, support vector machines, clustering, anomaly detection, recommender systems and general advice for applying machine learning techniques.

The Matlab/Octave programming exercises are well-designed, and for someone who didn't know Octave before, learning this was a bonus.Compared with Hinton's 2012 course, Ng's is better for beginners, covers slightly different topics (Support Vector Machines, but not Boltzmann Machines, Belief Nets, and Autoencoders), and makes a knowledge of calculus optional (though it's more interesting if you have this - especially chain rule).

The course broadly covers all of the major areas of machine learning -- linear and logistic regression, neural networks, support vector machines, clustering, dimensionality reduction and principal component analysis, anomaly detection, and recommender systems.

Read more

covers a wide range in 12 reviews

Covers a wide range of topics with great explanations for the theory and good practical programming excersizes.

Andrew Ng is an excellent teacher and covers a wide range of relevant material, with very topical examples and applications.

The course covers a wide range of techniques, definitely recommended.

It covers a wide range of contents and focuses on application of those machine learning methods.

Covers a wide range of machine learning applications and methods.

Es de lo mejor A great course that covers a wide range of machine learning techniques.

It's a good introduction - not too complicated and covers a wide range of topics.

不错,系统地学习了机器学习的内容,讲的深入浅出,很实用。 très bon professeur et lecture This free course, which is available for certificate now is well designed and covers a wide range of topics.

Covers a wide range of topics.

Covers a wide range of machine learning algorithms and critically, teaches how to measure and evaluate different approaches.

Overall the course covers a wide range of topics.

Read more

inner workings in 17 reviews

For me, Machine learning had mostly been magic... until I enrolled in Andrew's course which dispelled the magic and explained the inner workings to someone who is new to this field as clearly as possible.

It's an amazing addition to the curricula especially after several years after my graduation.For some people might be too basic, for some others like me, it was precisely what I needed to know, to understand the inner workings, to be able to "tune up the car, and not just be a driver".

This course is excellent for understanding the inner workings of Machine learning.

Go deep into the fundamentals rather than just using scripts that you don't understand the inner workings of.

Andrew Ng found the perfect balance between explaining seemingly complicated concepts in simple terms, while at the same time still giving enough mathematical detail that one gets to understand the "inner workings" of the machine learning techniques presented in this course.

Really descriptive and thorough, teaching you the fundamentals and inner workings behind machine learning and more advanced topics as well.

The course doesn't just treat the algorithms as black boxes, but rather explains the mathematics and inner workings of them.

It would have been interesting to hear a little more about the inner workings of SVMs, this topic was handled a little too short for my taste.

Andrew Ng makes a great job explaining the concepts and especially giving us an intuition about the inner workings of the presented methods and algorithms during the course.Even more importantly he goes beyond the theory and gives us very useful guidelines to apply in production, when it is time to use machine learning to tackle real life problems.Last but not least Andrew is a charismatic teacher which transmits his passion and fun for the topic.

Great course where you learn the fundamentals, and inner workings of machine learning algorithms.

A great course for anyone who needs to learn the inner workings and the essence of machine learning.

I especially liked how the algorithms were derived and how it is possible to get a pretty good understanding of their inner workings.

Especially clear and thorough on concepts and inner workings of ML algorithms.

I would highly recommend this class to anyone who is interested in the inner workings behind the recent craze of ML.

Read more

without getting bogged down in 10 reviews

This course does a great job of surveying some of the most popular and useful machine learning algorithms and digs into the math enough to give good intuition without getting bogged down in the details.

I really enjoyed Andrew Ng's teaching style, and even though I have a fairly extensive background in math, I appreciated his emphasis on applying these techniques without getting bogged down in proofs, etc.

The difficulty often lies finding that boundary — the boundary where the complexity of a computation or a problem or a strategy can be abstracted out (with a black-box, or an analogy) and a student can make progress in thinking about the problem without getting bogged down.

Great teaching style , Presentation is lucid, Assignments are at right difficulty level for the beginners to get an under the hood understanding without getting bogged down by the superfluous details.

If you want to apply machine learning for your work or project without getting bogged down by the intricate details (though it does provide snippets of the technical terms which you can learn it yourself later), then this is the course for you.

Ng, through this courses, demystifies machine learning and imparts useful skills without getting bogged down by the mathematics - if you're math-shy.

It goes into an appropriate amount of detail without getting bogged down by esoteric mathematics.

It gets into the background maths without getting bogged down with it.

The course instructor hinted at a lot of the theory that underlay the topics without getting bogged down in the rigor and thus gave a very good pragmatic and intuitive introduction to the material.

Read more

right balance between theory in 9 reviews

Fast paced and the right balance between theory and useful applications.

A wonderfully well structured course with the right balance between theory, insights and examples of applications of the learning algorithms.

Very good course, solid structure and right balance between theory and practice.Recommended!

The materials prepared by Prof Andrew and his team strikes a right balance between theory and practice.

Very good introduction to Machine Learning with the right balance between Theory (Maths) & PracticeWhat could have been even more awesome: Use R for assignments, An Overview of the other main ML Algorithms (decision trees, rules...) and a Summary of conditions where to use or not each category of algorithmBig thanks Andrew!!

Excellent course Great course, useful content, right balance between theory and practical exercises, taught by Prof Andrew Ng who is exceptional not only in teaching effectively the topics covered in the course but also passing on to you his passion for machine learning.

I am really enjoying this course!Right balance between theory and programming!

Great tutorial exercise Very high level, skips over a lot of the harder math, but gives a good overview over most of the main algorithms and talks a lot about helpful methods to improve them/use your time wisely Just about the right balance between theory and application.

Read more

silicon valley in 14 reviews

As Andrew Ng says, after the course you will know more than most engineers in Silicon Valley.

It also explains very well how to work with different ML algorithms, how to monitor they are "learning well", and how to fine-tune their parameters or tweak the inputs, in order to gain better results.The course ends with assuring students that their skills are "expert-level" and they are ready to do amazing things in Silicon Valley.

Although I would have preferred a little more focus on the theory/mathematics underlying the algorithms, as opposed to silicon valley application tricks.

i particularly liked you silicon valley insights!

He is also very encouraging with statements that "By now you know more than the average Silicon Valley Engineer"... Bottom line is this course helps new bies grasp and love machine learning This course filled in many gaps in my knowledge and was presented at just the right level.

Prof. Ng has prepared many materials, including application cases in Silicon Valley, to illustrate the essence of machine learning.

In addition to the subject matter itself, it is interesting to learn that these are some of the key methods used in Silicon Valley.If you have a strong mathematical background, some of the early lectures are too slow.

Several times throughout the course, Andrew mentioned that learning the material presented in the course would put you above most ML users in Silicon Valley.

Another thing that I found great is the industry reference made, and that the same technology is being used in the Silicon Valley.

Its also quite annoying, how the lecturer calls anyone with a basic understanding of a mathematical field an expert while simultaneously claiming that most of the people in the silicon valley are clueless.In summary, its a good course for high schoolers, but anyone with a little mathematical background should rather spend his time/money with an other course.

(I had sold all chickens I raised in the vallage, and purchased a flight ticket to silicon valley so as to find a job.

See you in Silicon valley> Nice experiences here to have learnt a lot of knowledge about machine learning.

I also took particular issue with Andrew Ng's smug remarks about how anyone who takes this course is now "an expert in machine learning" or that 3 weeks in you already know more than many people doing ML in silicon valley.

I'm literally in silicon valley and there are no ML jobs to be had for anyone who doesn't have a dedicated degree and/or at least 3 years experience with whatever software they are using.

Read more


An overview of related careers and their average salaries in the US. Bars indicate income percentile.

Research Scientist-Machine Learning $55k

Cloud Architect - Azure / Machine Learning $75k

Watson Machine Learning Engineer $81k

Machine Learning Software Developer $103k

Software Engineer (Machine Learning) $116k

Applied Scientist, Machine Learning $130k

Autonomy and Machine Learning Solutions Architect $131k

Applied Scientist - Machine Learning -... $136k


Machine Learning Engineer 2 $161k

Machine Learning Scientist Manager $170k

Machine Learning Scientist, Personalization $213k


Sorted by most helpful reviews first

Guest says:

Amazing. I learned regression (linear and logistic), k-means clustering, and principal components analysis in uni, but seeing it in an ML context reinforced the grasp I have over those topics. Neural networks were entirely new to me and by far one of the most interesting concepts I've learned from anywhere!

Guest says:

this is a very good introduction to machine learning, i have already used it on two projects about for image recognition and customer service logs from company server. if you google it, you can also find many machine learning libraries for python, C++, etc. but in my opinion you can use those more effectively if you know what is happening behind the scene.

Write a review

Your opinion matters. Tell us what you think.



Stanford University

Rating 4.9 based on 28,977 ratings
Length 12 weeks
Starts May 11 (3 weeks ago)
Cost $79
From Stanford University via Coursera
Instructor Andrew Ng
Download Videos On all desktop and mobile devices
Language English
Subjects Programming Data Science
Tags Computer Science Data Analysis & Statistics Data Science Machine Learning

Similar Courses

Sorted by relevance

Like this course?

Here's what to do next:

  • Save this course for later
  • Get more details from the course provider
  • Enroll in this course
Enroll Now