Chevron Left
Zurück zu Bayesian Methods for Machine Learning

Kursteilnehmer-Bewertung und -Feedback für Bayesian Methods for Machine Learning von National Research University Higher School of Economics

4.6
412 Bewertungen
111 Bewertungen

Über den Kurs

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...

Top-Bewertungen

JG

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.

LB

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.

Filtern nach:

76 - 100 von 106 Bewertungen für Bayesian Methods for Machine Learning

von Bart-Jan V

Nov 23, 2018

Great course, great material, though difficult to follow a non native English speaker being non-english myself. Though the instructors know what they are talking about, they don't tell it in their own words but rather seem to have practiced their text.

Another important point is that it took me a lot of time to follow (pre)calculus and probability theory courses, to be able to understand this course. The course was a nice motivation to do that. I'm glad I did, because now I can understand and use VAE's and bayesian optimization (and some other useful stuff)

von Ishaan B

Nov 28, 2018

The content+course structure was phenomenal. The assignment environment setup was a bit cumbersome at times, but the level of difficulty in the assignments really solidified the understanding of the course material.

von Milos V

Jan 08, 2019

As PhD in physics I found lecture super-boring (too much theory and derivation) and irrelevant to the practical assignment. On the other hand, most of practical assignments are explained very pedagogical manner (except week 5!). As for the first course - I would recommend more code-related lectures.

von 魏力

Jul 21, 2018

Good course. But some suggestions: topic about variational inference or variational EM in theory is quite tough, better to have equivalent level of assignment for better practical understanding. Personally, I feel VAE is a very simplified application case.

von Maury S

Aug 22, 2018

Excellent, detailed content for people wanting to understand variational methods for machine learning. Fairly high degree of math and statistics required as a prerequisite, as well as moderate ability as a Python programmer. Does not get 5 stars because some of the assignments had confusing instructions, and availability of instructors and others to asnwer questions was poor.

von Hugo R C R

Jun 19, 2018

It probably offers the most comprehensive overview of Bayesian methods online. However, it would be nice these methods translate into practical data science problems found in the industry.

von Chiang y

Jun 04, 2018

We may need more help for homework format or quiz answer format. It took me lots time for solving it.

von 冯迪(Feng D

Feb 26, 2018

The materials of this lecture are awesome. Very useful! However, the introduction of project assignments are very confusing, especially the final project. It took me hours to understand what the task is really about, and what should we really do.

von Olaf W

Jun 26, 2018

Great class. Well presented material. Sometimes the path from introduction to advanced material could use a few steps in between.

von Mauro D S

Jul 03, 2018

Hard material, but very well explained. The peer-review exercises are interesting as well, but if the reviewer does not understand the material, I wonder how useful they are. Open research question I guess (i.e. how to make sure the student reviewer understands what he is reviewing-are there any baseline reviews established a student should go through first?)!

von Guy K

Mar 19, 2018

a very important material is covered in a clear manner.

some of the labs could have been more effective (e.g. avoid unnecessary mixing between tensorflow and Keras)

Strongly recommended course ! great curriculum !

von Alexander E

Jun 02, 2018

Excellent material! I got new very useful knowledge. I really like the final project. Although course design is not perfect. It would be great to have additional content (links or documents), lectures are not enough to pass the tests. Also some assigments have issues (code and grader errors).

von 洪贤斌

Aug 30, 2018

Good course but a bit difficult and the peer review is helpless

von Joris D

Jul 17, 2018

I can not recommend this course highly enough. Unfortunately I can't give it 5 stars since some of the computer assignments were outdated with respect to the tools they utilize (e.g. arguments in the assignments not existing anymore). Still, let that not discourage you. If you ever mentally disconnect when people start talking about Gibbs sampling, mean field approximations, intractable variational lower bounds, or other big fancy words, this is definitely the course for you. You'll discover that all these things are actually quite straightforward.

von MASSON

Apr 06, 2019

Good course.

Too much theory, not enough practice

von Mehrdad S

Sep 03, 2019

This is a great course for some of the advance topics in Baysian ML. The course starts off great and provides great explanation of the basic topics such as Conjugate, EM algorithm, etc. The related HW are also intelligently designed and fun to solve. But, as it reaches the weeks 5 and 6, things starts to fall apart and the materials are not presented and explained in the best possible way. I think the instructors try to teach many topics which requires a little bit of patience in a short amount of time. Overall, I believe its a course worthy of try, certainly provides great exposure to some of the advance topics but requires further follow ups and studies to completely digest all the materials.

von Pengchong L

Aug 28, 2018

Not very well prepared. Contents are dry and not well illustrated. Failed to explain points that are made in the videos. The lecturers are reading from scripts and look very nervous.

von Artem E

Jun 03, 2018

Not so good as I thought. Some times is too complicated and dry. Need more balance. I hope, that guys can better. But I want to say thanks to authors. You did a great job! Good luck.

von Lavinia T

Jan 29, 2018

The trainer's English is not very good, and the explanations provided are insufficient.

von Siwei Y

Feb 20, 2018

给三星是因为所选的 TOPICS 很好, 真的很好。但是,说到老师的讲解,就真的不敢恭维了。从逻辑性到流畅性都让人捏把汗啊。希望改进。

von Beibit

Jun 27, 2019

As the description suggests this course is very advanced and math heavy.

von 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.

von hyunseung2 c

Sep 19, 2019

ㅁㄴㄹ

von Ahmad

Jan 16, 2019

Not structured well

von Gourab C

Jun 26, 2018

I felt the explanations too mechanical and in between they skipped a lot of concepts and explanations.