返回到 Bayesian Methods for Machine Learning

4.6

星

530 個評分

•

155 條評論

People apply Bayesian methods in many areas: from game development to drug discovery. They give superpowers to many machine learning algorithms: handling missing data, extracting much more information from small datasets. Bayesian methods also allow us to estimate uncertainty in predictions, which is a desirable feature for fields like medicine.
When applied to deep learning, Bayesian methods allow you to compress your models a hundred folds, and automatically tune hyperparameters, saving your time and money.
In six weeks we will discuss the basics of Bayesian methods: from how to define a probabilistic model to how to make predictions from it. We will see how one can automate this workflow and how to speed it up using some advanced techniques.
We will also see applications of Bayesian methods to deep learning and how to generate new images with it. We will see how new drugs that cure severe diseases be found with Bayesian methods.
Do you have technical problems? Write to us: coursera@hse.ru...

Nov 18, 2017

This course is little difficult. But I could find very helpful.\n\nAlso, I didn't find better course on Bayesian anywhere on the net. So I will recommend this if anyone wants to die into bayesian.

Jun 07, 2019

Excellent course! The perfect balance of clear and relevant material and challenging but reasonable exercises. My only critique would be that one of the lecturers sounds very sleepy.

篩選依據：

創建者 Голубев К О

•Oct 19, 2018

Great course with fine lecturers and deep immersion in Bayesian methods

創建者 Dongxiao Z

•Oct 11, 2018

Learned a lot from this course. Thanks!

創建者 Alexander R

•Nov 12, 2018

super helpful and very applicable!

創建者 Anmol G

•Dec 06, 2018

One of the best in-depth course.

創建者 SagarSrinivas

•Sep 29, 2018

Awesome. Worth it!

創建者 Max P Z

•Apr 02, 2018

Tough but useful!

創建者 Ertan T

•Apr 26, 2018

Superb Course

創建者 Daniel T

•Aug 06, 2019

The material is good and a lot of effort went into designing this course. Nonetheless, it feels neglected and could use an update.

The presentations are somewhat muddled by notational abuse. Indeed, it's customary to shorthand every distribution as "p" and let the arguments remind you which variable it came from, e.g, p(x|y) is conditional density of variable "X" at x given that "Y" = y. But then "p(a|b)" could be a completely different function corresponding to random variables "A" and "B"; however, you could have a=x and y=b as vectors which amplifies confusion... And when many variables with different ranges are involved and there's no consistency between labels for the variables and labels for their values, one has to spend extra time deciphering the material. Keeping track of the random variables and adopting a more suggestive notation would go a long way. Also, in Bayesian context it helps to avoid the word "parameter" (other than hyper-parameter, maybe), e.g., the weights "w" themselves are just values of a random variable, which is no different than the data generating process or the latent variables.

The programming assignments contain a lot of missing or inconsistent instructions. Be prepared to sift through the forums to find what is really expected or how to fix the issues in the supplied code.

Overall, I get the impression the course is now maintained by the students. It would be nice to see a revision from the instructors.

創建者 Jean M A S

•Jun 03, 2018

Well, this course is really good, very demanding, and rigorous. The main disadvantage is the forum. Chances are that nobody will answer your questions, so be prepared to have a raw experience of learning. But if you are serious, you will eventually finish the course, and learn a lot.

創建者 Maciej

•Mar 24, 2019

Overall it's good. My problem is that most of this material is better suited to lecture notes and not a video. They're forcing it into a video since it's coursera. Couldn't get through a lot of the lectures, used a textbook instead.

創建者 张学立

•Nov 08, 2017

it seems that the prof didn't prepare the course well

創建者 Dizhao J

•Aug 08, 2018

very bad Interpretation

創建者 Novin S

•Feb 03, 2020

I really enjoyed taking this course. The quality of lectures and material were really good, and it was advanced topics as promised. The theories were addressed sufficiently with examples from the real world which made the course not only theoretically interesting but also practically applicable and useful. There have been tiny issues here and there, either during the homework assignments or the material but I hope those will be fixed together with new updates to the course to keep it up to date with the state of the art of the research in the field of Machine Learning.

創建者 Mayukh S

•May 11, 2020

This is one of the best courses I've come across in coursera. The topics are covered in detail. The best part are the proof's for every algorithm they use. This helps in developing useful insights which helps in using these algos for other problems. The assignments and quizzes are challenging. They require the learner to read documentations of libraries and try to come to a solution. Everything is not provided as other courses which is a very good things as this is a advanced course and requires learners to put that extra effort. I would highly recommend this course.

創建者 Martin K

•Mar 16, 2018

The course material is very well prepared and self-contained. Derivation of relevant mathematical formulas is done in great detail which was really helpful. If you've read books like Murphy's "Machine Learning - A Probabilistic Perspective" or Bishop's "Pattern Recognition and Machine Learning" then this course should be easy to follow. If not, it is helpful to have one of these books at hand to get a better understanding, as some topics are presented in a rather condensed form. Thanks to the lecturers for preparing this great course. I can highly recommend it!

創建者 Jordi W

•Feb 28, 2019

This is a challenging course, but well worth it! One needs to be able to manage both the lecture content and the practical side of the course, namely the Python modules/environment. The Python ecosystem is developping fast and some modules changed since the assignments have been created. This means that you need to be able handle deprecations within Python modules and your own Python environment if needed. But this is an advanced course, so I think that is fine. Things have been made easier now that the course creaters have moved assignments to Colaboratory.

創建者 Erwin P

•Mar 17, 2019

This course provides a comprehensive overview how Bayes stats can be used in ML. I'm better able to value the different concepts like EM, GP and VAE and put them into perspective. Depending on you previous math and stats skills the assignments can be challenging and it took me some stamina to complete. The "Russian English" is sometimes a bit of a hurdle when watching the videos, but you get used to it. The concepts are well explained and the references to the additional materials useful.

創建者 Marcos C

•Oct 17, 2019

This course was a fantastic intro to modern Bayesian methods. I particularly liked the references to relevant papers and the useful programming assignments.

The only negative I would say with this course (and all the courses in the specialisation) is that there is usually not enough density of people taking the course so the peer graded assignments take ages to be graded. I would recommend that these bits are made optional and don't count towards the final grade.

創建者 Alya S

•Oct 29, 2019

Very well structured and delivered course. The explanations are generally easy to follow and reproduce. Highly enjoyable and instructive. Assignments are relevant. It would have been great to have an assignment for the Dirichlet Allocation this would have improved the overall understanding of the algorithm. Overall very satisfied I took this lecture. Thanks very much to the lecturers.

創建者 VAISAKH S

•Feb 29, 2020

Amazing course with the right balance of mathematics and practicals... I would say a bit more mathy... I took around 10 weeks to complete this 6 week course since I was new to this area... But I would say my understanding has grown to such an extent that I can easily read papers in this area and make my own derivations for approximate inference... Thanks you guys

創建者 Bob F

•Mar 11, 2018

This class provided excellent lectures and very instructive programming assignments. I don't think that the material covered is available in any other MOOC. This class is among the very best I've taken, which is saying a lot because they have to compete with Andrew Ng, Geoff Hinton, and Chris Manning - just to mention few! Thanks for all the great work!

創建者 Ayush T

•Aug 24, 2019

It is undoubtedly one of the best course on Coursera that I've come across. This is really well taught and there is a good balance between the theoretical and the practical aspect of the Bayesian Machine Learning. This course is must-do for those who want to do some good projects in the field of Bayesian Deep Learning which is currently a hot topic now.

創建者 Pablo V I

•Jan 13, 2020

One of the most technicals and high-quality MOOCs I have completed. You need prior knowledge about machine learning and bayesian statistics to complete the assignments.

I highly recommend this course for people working in the industry or researchers. If you are looking for a challenging course, this is your choice.

創建者 Kuldeep J

•Apr 04, 2019

Various advanced Machine Learning topics like Bayesian interpretation techniques, probabilistic modelling, variational auto encoders, etc. have been explained in a very intuitive and simple manner. Then the assignments are well designed to make sure one is able to work on the existing packages available.

創建者 Igor B

•Apr 18, 2019

A wonderful course to improve the theoretical understanding of machine learning and recap probability theory. The lecturers did their best to drag the listener through the math of the EM algorithm and more. The transition to Google Colab indeed simplified online work with Jupyter notebooks.