Neurobayesian models 2019 — различия между версиями

Материал из Wiki - Факультет компьютерных наук
Перейти к: навигация, поиск
Строка 30: Строка 30:
 
===Assignments  ===
 
===Assignments  ===
 
There are three practical assignments. Usually, they are submitted in  [https://anytask.org/course/444 anytask]. To get the invite please write to course mail. The site has only Russian interface so the foreign student can submit to course mail. In this case, the subject line should consist of your name, surname and assignment number.
 
There are three practical assignments. Usually, they are submitted in  [https://anytask.org/course/444 anytask]. To get the invite please write to course mail. The site has only Russian interface so the foreign student can submit to course mail. In this case, the subject line should consist of your name, surname and assignment number.
 +
 
All assignments should be coded in Python 3.
 
All assignments should be coded in Python 3.
 +
 
Students have to complete all assignments by themselves. If the solution was discussed together, or any third-party codes and materials were used, then this should be written in the report. Otherwise, “similar” solution would be considered as plagiarism and all involved students (including those who share his solution) will be severely punished.
 
Students have to complete all assignments by themselves. If the solution was discussed together, or any third-party codes and materials were used, then this should be written in the report. Otherwise, “similar” solution would be considered as plagiarism and all involved students (including those who share his solution) will be severely punished.
 +
 
Assignments are scored up to 10 points. Each practical assignment has a deadline, a penalty is charged in the amount of 0.3 points for each day of delay, but in total not more than 6 points. Some assignments may contain bonus part.
 
Assignments are scored up to 10 points. Each practical assignment has a deadline, a penalty is charged in the amount of 0.3 points for each day of delay, but in total not more than 6 points. Some assignments may contain bonus part.
  

Версия 22:21, 6 февраля 2019

The page is not ready yet!

Lector: Dmitry Vetrov

Tutors: Alexander Grishin, Kirill Struminsky, Dmitry Molchanov, Kirill Neklyudov, Artem Sobolev, Arsenii Ashukha, Oleg Ivanov, Ekaterina Lobacheva.

Contacts: All the questions should be addressed to bayesml@gmail.com. Theme of any letter must contain the following tag: [HSE NBM19]. Letters without the tag will be most probably lost in the inbox.

We also have a chat in Telegram (link to it was sent to the group email). It's main language is Russian, but all the questions in English will be answered in English. All important news will be announced in English in the chat and also sent to the group e-mail.

Course description

This course is devoted to Bayesian reasoning in application to deep learning models. Attendees would learn how to use probabilistic modeling to construct neural generative and discriminative models, how to use the paradigm of generative adversarial networks to perform approximate Bayesian inference and how to model the uncertainty about the weights of neural networks. Selected open problems in the field of deep learning would also be discussed. The practical assignments will cover implementation of several modern Bayesian deep learning models.

Course syllabus

News

Grading System

The assessment consist of 3 practical assignments and a final oral exam. Practical assignments consist in programming some models/methods from the course in Python and analysing their behavior: VAE, Normalizing flows, Sparse Variational Dropout. At the final exam students have to demonstrate knowledge of the material covered during the entire course.

Final course grade is obtained from the following formula:

О_final = 0,7 * О_cumulative + 0,3 * О_exam,

where О_cumulative is an average grade for the practical assignments.

All grades are in ten-point grading scale. If О_cumulative or О_final has a fractional part greater or equal than 0.5 then it is rounded up.

Assignments

There are three practical assignments. Usually, they are submitted in anytask. To get the invite please write to course mail. The site has only Russian interface so the foreign student can submit to course mail. In this case, the subject line should consist of your name, surname and assignment number.

All assignments should be coded in Python 3.

Students have to complete all assignments by themselves. If the solution was discussed together, or any third-party codes and materials were used, then this should be written in the report. Otherwise, “similar” solution would be considered as plagiarism and all involved students (including those who share his solution) will be severely punished.

Assignments are scored up to 10 points. Each practical assignment has a deadline, a penalty is charged in the amount of 0.3 points for each day of delay, but in total not more than 6 points. Some assignments may contain bonus part.

Approximate dates for homework assignments (they can change!): TBA

At the end of the module before the exam there will be a hard deadline for all assignments! Exact date will be announced later.

Exam

TBA

Course Plan

Занятие Дата Название Материалы
1 24 September Lecture: Stochastic Variational Inference article
2 31 September Seminar: Application of SVI to Latent Dirichlet Allocation model article
Lecture: Doubly Stochastic Variational Inference TBA
3 7 October Seminar: Doubly Stochastic Variational Inference TBA
Lecture: Variational autoencoders (VAE) and normalizing flows (NF) VAE article, NF article
3 14 October Seminar: Importance Weighted Autoencoders + more complex NF TBA
Lecture: Density ratio estimation + alpha-GAN article
3 21 October Seminar: f-GAN article
Lecture: Bayesian neural networks article, article, article
3 28 October Seminar: Local reparametrization trick article
Lecture: Bayesian compression of neural networks article, article
3 7 November Seminar: Deep Marcov chain Monte Carlo (MCMC) article article
Lecture: Variance Reduction article
3 14 November Seminar: Discrete latent variables article article article
Lecture: Semi-implicit variational inference article, article
3 21 November Seminar: VampPrior article, article

Reading List

  • Murphy K.P. Machine Learning: A Probabilistic Perspective. The MIT Press, 2012.
  • Bishop C.M. Pattern Recognition and Machine Learning. Springer, 2006.
  • Mackay D.J.C. Information Theory, Inference, and Learning Algorithms. Cambridge University Press, 2003.
  • Ian Goodfellow, Yoshua Bengio & Aaron Courville. Deep Learning. MIT Press, 2016.

Useful links

[The same course in Russian at MSU] (contains more materials in Russian).
BayesGroup page.