Neural Networks for Machine Learning

Provided by:
8/10 stars
based on  15 reviews
Provided by:
Cost FREE
Start Date On demand

Course Details

Cost

FREE

Upcoming Schedule

  • On demand

Course Provider

Coursera online courses
Coursera's online classes are designed to help students achieve mastery over course material. Some of the best professors in the world - like neurobiology professor and author Peggy Mason from the University of Chicago, and computer science professor and Folding@Home director Vijay Pande - will supplement your knowledge through video lectures. They will also provide challenging assessments, interactive exercises during each lesson, and the opportunity to use a mobile app to keep up with yo...
Coursera's online classes are designed to help students achieve mastery over course material. Some of the best professors in the world - like neurobiology professor and author Peggy Mason from the University of Chicago, and computer science professor and Folding@Home director Vijay Pande - will supplement your knowledge through video lectures. They will also provide challenging assessments, interactive exercises during each lesson, and the opportunity to use a mobile app to keep up with your coursework. Coursera also partners with the US State Department to create “learning hubs” around the world. Students can get internet access, take courses, and participate in weekly in-person study groups to make learning even more collaborative. Begin your journey into the mysteries of the human brain by taking courses in neuroscience. Learn how to navigate the data infrastructures that multinational corporations use when you discover the world of data analysis. Follow one of Coursera’s “Skill Tracks”. Or try any one of its more than 560 available courses to help you achieve your academic and professional goals.

Provider Subject Specialization
Humanities
Sciences & Technology
4680 reviews

Course Description

Learn about artificial neural networks and how they're being used for machine learning, as applied to speech and object recognition, image segmentation, modeling language and human motion, etc. We'll emphasize both the basic algorithms and the practical tricks needed to get them to work well.
Reviews 8/10 stars
15 Reviews for Neural Networks for Machine Learning

Ratings details

  • 5 stars
  • 4 stars
  • 3 stars
  • 2 stars
  • 1 stars
  • 5 stars
  • 4 stars
  • 3 stars
  • 2 stars
  • 1 stars
  • 5 stars
  • 4 stars
  • 3 stars
  • 2 stars
  • 1 stars

Rankings are based on a provider's overall CourseTalk score, which takes into account both average rating and number of ratings. Stars round to the nearest half.

Sort By
Anna Nachesa profile image
Anna Nachesa profile image
10/10 starsCompleted
  • 13 reviews
  • 13 completed
5 years, 11 months ago
As was already said on this site, this course is a great follow-up to the Machine Learning by Andrew Ng. The material is more advanced (specifically for me the Restricted Boltzmann Machines were a big surprise, I have never heard this term before) and you need to do a lot of own study to be successful in this course. The quality of the material, both the video and the lecture slides, was quite good. I only wish there were more links to the related research and/or books. Of course, this is one of the courses which might leave you with the idea of how little you know about the subject after all, but this is better than the illusion of full knowledge :) Prerequisites: some programming experience, math on the undergraduate level. Previous exposure to machine learning helps to lower the amount of new information and to focus on the novel ideas.
Was this review helpful? Yes4
 Flag
Student profile image
Student profile image

Student

10/10 starsTaking Now
2 years, 5 months ago
This course has great content ( class materials), although not always up to date. I have all the respect to Dr.Hinton, however, I found some his explanation were not on the key points. I don't like the design of the quizzes and the assignments. Put the errors in the quiz evaluation on the side, the questions themselves did not help understanding the topic but quite tedious experiments.
Was this review helpful? Yes0
 Flag
M. Rez. profile image
M. Rez. profile image
4/10 starsCompleted
  • 1 review
  • 1 completed
2 years, 6 months ago
The course covers most of the important materials that I expect to be covered in the topic, but it is being presented in a very vague and confusing way. The intuitions are not conveyed the way I was expecting them to be. I have seen better content out there in youtube.
Was this review helpful? Yes0
 Flag
Student profile image
Student profile image

Student

10/10 starsTaking Now
2 years, 7 months ago
Hinton is clear, concise, brilliant. Note that this is a difficult course, I'm sure that is the cause of any downvoting. You'll be expected to use your initiative to bone up on Bayesian Inference and the necessary math.
Was this review helpful? Yes0
 Flag
Student profile image
Student profile image

Student

10/10 starsTaking Now
2 years, 8 months ago
Hinton is brilliant, concise, succint, clear. Every word / diagram is consciously architected. I'm surprised anyone is downvoting, maybe it is because the material is more difficult than they expected? Certainly need to rewatch certain videos multiple times.
Was this review helpful? Yes0
 Flag
Arnaud Sors profile image
Arnaud Sors profile image
10/10 starsCompleted
  • 1 review
  • 1 completed
3 years, 6 months ago
Absolutely brilliant course. I am in awe of how Mr Hinton managed to wrap up this comprehensive overview of the field. This course requires the student to persevere and think by himself and therefore may not be ideal (or fun-maximizing) for a very first intro to NNs (for this I would advise Andrew Ng's online tutorial on UFLDL website) but has the big merit of extensively and effectively covering most NN techniques, whilst also providing historical insight into how and why these techniques were subsequently invented. Very useful. Je tire mon chapeau.
Was this review helpful? Yes2
 Flag
Zbyněk Zajíc profile image
Zbyněk Zajíc profile image
10/10 starsCompleted
  • 18 reviews
  • 18 completed
3 years, 7 months ago
Very good overview about the neural networks and not only the basics, Fully recommended to the students with a basics of AI. I didn't do the homeworks, so I'm rating only the lectures.
Was this review helpful? Yes0
 Flag
 profile image
 profile image

6/10 starsDropped
  • 1 review
  • 0 completed
3 years, 10 months ago
Hinton is a great researcher but the delivery of this course was unorganized and leaves many holes unfilled. Furthermore the course is rattled with messed up font and incorrect placing of quizzes.
Was this review helpful? Yes2
 Flag
Student profile image
Student profile image

Student

2/10 starsCompleted
3 years, 10 months ago
I have the utmost respect for Professor Hinton as a scientist, but I was very disappointed by this class when I took it. It's fairly good on a superficial level (reasonable topic selection + obvious mastery of the subject from the theacher), but the complete lack of details and insight makes the whole material useless for anyone trying to throughly comprehend / replicate the analysed topics. The classic approach is something along the lines of 'well, I've put 3 layers, the first one with 6 neurons, the second one with 100 and the third one with 8, trained it and obtained this pattern': no information about how those values (number of layers and their size) were picked, nor about how those pattern emerged. The whole class is weirdly unbalanced, with this huge focus on deriving each error function + these out-of-the-blue experimental results. It's like explaining the geometry of the cylinder for about one hour and then briefly mentioni... I have the utmost respect for Professor Hinton as a scientist, but I was very disappointed by this class when I took it. It's fairly good on a superficial level (reasonable topic selection + obvious mastery of the subject from the theacher), but the complete lack of details and insight makes the whole material useless for anyone trying to throughly comprehend / replicate the analysed topics. The classic approach is something along the lines of 'well, I've put 3 layers, the first one with 6 neurons, the second one with 100 and the third one with 8, trained it and obtained this pattern': no information about how those values (number of layers and their size) were picked, nor about how those pattern emerged. The whole class is weirdly unbalanced, with this huge focus on deriving each error function + these out-of-the-blue experimental results. It's like explaining the geometry of the cylinder for about one hour and then briefly mentioning 'oh yeah, we use those as wheels, to make vehicles move'.
Was this review helpful? Yes2
 Flag
Dan profile image
Dan profile image
10/10 starsCompleted
  • 9 reviews
  • 9 completed
5 years, 3 months ago
Wow, one of a few truly great classes I've taken on Coursera. Also the hardest from the math point of view and assignments. The quizzes had some hiccups (e.g. some questions were initially ambiguous, but later corrected) and were hard and required lots of thought, but I really learned a lot from them, especially about how to get the learning formulas for networks with arbitrary sequences of different types of layers. Good treatment of both basic and advanced neural network topics, with great insights about how different areas are related and about practical issues and solutions when working with neural networks (mini-batches, many different methods to avoid overfitting, etc.). The lectures and slides were not perfect as at times they seemed to jump into new topics without much preparation or the jargon changed without explanation from one topic to the next... but they were small issues and overall the lectures were great. Programming... Wow, one of a few truly great classes I've taken on Coursera. Also the hardest from the math point of view and assignments. The quizzes had some hiccups (e.g. some questions were initially ambiguous, but later corrected) and were hard and required lots of thought, but I really learned a lot from them, especially about how to get the learning formulas for networks with arbitrary sequences of different types of layers. Good treatment of both basic and advanced neural network topics, with great insights about how different areas are related and about practical issues and solutions when working with neural networks (mini-batches, many different methods to avoid overfitting, etc.). The lectures and slides were not perfect as at times they seemed to jump into new topics without much preparation or the jargon changed without explanation from one topic to the next... but they were small issues and overall the lectures were great. Programming assignments were great, although runtime was long (good luck trying to do it starting only a couple of hours before the deadline) and I wish there was deep network practical coverage. It's great to see how much neural network have progressed since I studied the basics at the university long ago. The class gives a quick overview of great application results of the last years for deep networks, and points to some good papers for those interested in learning more details. If this class is offered again (and if you have basic math skills like derivation), don't miss it!
Was this review helpful? Yes0
 Flag
Trevor profile image
Trevor profile image
8/10 starsCompleted
  • 4 reviews
  • 4 completed
5 years, 7 months ago
A great follow on course after doing the Andrew Ng Machine Learning. Some very interesting real world insights and new ideas.
Was this review helpful? Yes0
 Flag
dan profile image
dan profile image
5/10 starsTaking Now
  • 2 reviews
  • 1 completed
6 years, 4 months ago
I love the material! The lectures are good, but this course has some serious problems. There are bugs in the slides and the quizzes, causing formulas to show up as things like "Undefined control sequence \dfrac" or " Misplaced \", making you guess at the part of the formula that's missing. Despite having a decent background in computational learning theory and a bit of background in ML, I found myself having to crack open a book to be able to do a lot of the quizzes. The material for many questions is only covered tangentially in lecture, if at all. I can only imagine how hard it would be with no background (the course material pointedly did not list ML background as being required). If you go to the coursera forums, you can see multiple highly upvoted threads complaining that the lectures simply don't cover the material needed for the quizzes. The staff have responded by saying they're thankful for the feedback, and will try to make... I love the material! The lectures are good, but this course has some serious problems. There are bugs in the slides and the quizzes, causing formulas to show up as things like "Undefined control sequence \dfrac" or " Misplaced \", making you guess at the part of the formula that's missing. Despite having a decent background in computational learning theory and a bit of background in ML, I found myself having to crack open a book to be able to do a lot of the quizzes. The material for many questions is only covered tangentially in lecture, if at all. I can only imagine how hard it would be with no background (the course material pointedly did not list ML background as being required). If you go to the coursera forums, you can see multiple highly upvoted threads complaining that the lectures simply don't cover the material needed for the quizzes. The staff have responded by saying they're thankful for the feedback, and will try to make things better next time; here's one exact quote: "Thanks for the feedback everyone (including everyone who votes for feedback threads). We're working out the right balance as we go, and I think we'll have it down better next time we run the course. I agree that there've been some real problems." It's great that they're acknowledging the problems and will try to fix them, but that doesn't help anyone currently taking the course. To summarize, the lectures are great, except for some bugs in the slides, and the overall course would be good for someone with a decent ML background. However, it's billed as a course for people that don't have that background; they say they'll try to make it better next time, but this run of the course has some problems.
Was this review helpful? Yes2
 Flag
Ben Haley profile image
Ben Haley profile image
9/10 starsTaking Now
  • 5 reviews
  • 4 completed
6 years, 5 months ago
I've you've taken Andrew Ng's machine learning course, this provides a great followup. Geoffrey Hinton takes us deep into neural networks, well past the back propagation algorithm that Andrew left off with. The content is more math intensive than ml class and the quizes are more rigorous, but not by too much. Assignments are done in octave (an open source Matlab clone). I've seen some complaints in the forums about questions and assignments being vague and underspecified. I don't feel that way very often, but I do think the learning curve will be steep if you haven't had a decent ml exposure already. The exciting part about this course is the content. Geoffrey is pushing us right to the brink of what's possible in terms of object and speech recognition. The deep neural networks that he is building too are really cutting edge. Over the last 4 years we've seen them make significant advances in many areas of machine learning. Geoffrey i... I've you've taken Andrew Ng's machine learning course, this provides a great followup. Geoffrey Hinton takes us deep into neural networks, well past the back propagation algorithm that Andrew left off with. The content is more math intensive than ml class and the quizes are more rigorous, but not by too much. Assignments are done in octave (an open source Matlab clone). I've seen some complaints in the forums about questions and assignments being vague and underspecified. I don't feel that way very often, but I do think the learning curve will be steep if you haven't had a decent ml exposure already. The exciting part about this course is the content. Geoffrey is pushing us right to the brink of what's possible in terms of object and speech recognition. The deep neural networks that he is building too are really cutting edge. Over the last 4 years we've seen them make significant advances in many areas of machine learning. Geoffrey is a master of the field which means that he combines technical expertise with a deep knowledge of how these systems work. If you want to contribute to the future of advanced machine learning, this class is a great launching point.
Was this review helpful? Yes2
 Flag
Robert Komartin profile image
Robert Komartin profile image
9/10 starsDropped
  • 19 reviews
  • 16 completed
6 years, 4 months ago
I think it's only fair to start by saying this is one of the courses I've not completed. I found the depth of the topics a bit too much for an amateur like me (even after successfully completing Andrew Ng's ML course). But give to Caesar what is Caesar's - Mr. Hinton is a true Lord of the Neural Networks - the not only masters the depths of the field, but he also does this with grace and good humor - thank you for the great experience!
Was this review helpful? Yes0
 Flag
Thomas Johnson profile image
Thomas Johnson profile image
8/10 starsCompleted
  • 3 reviews
  • 3 completed
6 years, 4 months ago
Didn't do the homeworks, so I'm rating only the lectures, which I found to be great. Hinton is a great lecturer and discusses cutting-edge techniques. Hinton's dry wit makes many of the lectures extremely funny as well.
Was this review helpful? Yes0
 Flag

Rating Details


  • 5 stars
  • 4 stars
  • 3 stars
  • 2 stars
  • 1 stars
  • 5 stars
  • 4 stars
  • 3 stars
  • 2 stars
  • 1 stars
  • 5 stars
  • 4 stars
  • 3 stars
  • 2 stars
  • 1 stars

Rankings are based on a provider's overall CourseTalk score, which takes into account both average rating and number of ratings. Stars round to the nearest half.