Fork of Lempitsky DL for HSE master students
Lecture and seminar materials for each week is in ./week* folders
- Autumn'16 track finished, new track eta next autumn. Meanwhile, course materials will be reused for YSDA.
- Any technical issues, ideas, bugs in course materials, contribution ideas - add an issue
- 27.12 - autumn'16 track finished
- 23.12 - added some bonus links in most weeks for those mysterious people who lurk this repo from outside the HSE.
- 23.12 - pre-exam Q&A happened. Exam will happen next monday at 16.40, room 311
- 17.12 - added MIT just for kicks
- 16.12 - exam/project rules published - https://github.com/yandexdataschool/HSE_deeplearning/wiki/How-exam-works - ping us on which problems you prefer (see slack for reservations; will add table tomorrow)
- 16.12 - cheched up whatever you sent us before 16.00 PM 16.12. Some of your submissions haven't been graded yet? ping us asap.
- 16.12 - bayesian lecture happened
- 13.12 - A note to HSE students - final deadline is department-forced to happen on 26.12. We are [probably] terribly sorry. Please make sure that your points so far is at least 40 or will get there in time. We'll also have a practical exam, with tasks distributed next lecture. Details
- 13.12 - another total chechup wave (everything you sent us by 01:30 13.12.2016 should have been checked. If not, ping us/create issue)
- 9.12 - week11 uploaded
- 2.12 - updated info
- 2.12 - checked up all submissions sent before 5 AM (we hope). If yours is missing, PM or issue us.
- 2.12 - winter is coming... winder is here. Deadlines are coming!
- 25.11 - week9 by Arseny Ashukha on deep learning for sound processing
- 18.11 - week8 by Dmitry Ulyanov
- 15.11 - HW checkup wave. If you sent us your homework solution before 05:00 15.11.2016 but we never replied yet - go get us @slack or post an issue here - we'll help
- 13.11 - we fix__d__ week7 assignment and uploaded all the PDFs. Homework checkup wave is underway :)
- x.11 - another wave of homework checkups happened
- 3.11 - week7 will happen on 11.11.2016 the regular way (friday, 18-00, room 400)
- 3.11 - we checked up all homework assigments sent to us before 4:00 am 3.11.16. If you were for some reason left behind - write us (mail, slack, github issue, irl, ...) - we'll fix that.
- 3.11 - week6 homework announced
- 1.11 - Week6 lecture will occur on Wednesday, 2.11 at 18-00 at approximately the same room 400.
- 26.10 - Preliminary poll on next lecture date: http://doodle.com/poll/cvew2rkn9u9x5ept . Please participate ASAP.
- 24.10 - Checked up homeworks "so far". If you sent us any mail@homework before 4.20 24.10.16 and we haven't replied yet - PM us in slack or create github issue or just contact us IRL - we'll fix that.
- 21.10 - There will be no classes on the 28.10;
- 21.10 - week5 notebook is NOT homework assignment notebook. Homework and deadlines TBA
- 20.10 - we now have a
for you if you got to the seminar without a notebook (with a phone) or have some temporary technical issues. Binder only lasts 1 hour so please do not use it for homeworks.
- 20.10 - current status of course staff is "Адъ и израиль", so the new wave of homeworks will be checked with a short delay (hope to finish by the weekend. The lectures will proceed as planned.
- 17.10 - week4 partially completed seminar uploaded
- 17.10 - a way for students to reduce lateness penalty on their homework assignments announced
- 14.10 - week4 lecture, seminar and homework uploaded
- 12.10 - If you sent us anything before 6-00 AM (Moscow) 12.10.16 and still got no reply / no score here - contact us (slack, issue on github, anything) - there may be a problem with e-mail delivery.
- 12.10 - for those rare specimen who read official curriculum - we'll have have to reorder the curriculum. Advanced vision goes 2 weeks forward, advanced text gets 2 weeks sooner. The rest stays as planned.
- 7.10 - If you are considering project ideas, please contact us (slack >= mail) asap to know you exist. You don't have to know actual project topic - just tell us that you're there and you may be up to something.
- 7.10 - We've now got da feedback form - it's fully anonymous and you can send there whatever you won't send us via slack/mail.
- 7.10 - Week3 lecture notes and homework uploaded.
- 30.09 - Week2 homework (lasagne/cifar): Please do not forget to add deterministic=True for your neural network when computing accuracy (not when training)
- 30.09 - added week2 lecture and homework.
- 30.09 - published scoreboard
- 23.09 - added week1 materials and homework
- 20.09 - by default we meet on Friday at 18-00, room 400. if you cannot make it, please send us an e-mail (course mail) or PM me in slack ASAP - we have a second parallel available and a few other options.
- 20.09 - HW0 deadline was shifted 1 wek into the future. Rejoice!
- 20.09 - grading info added
- 15.09 - Doodle on when do we meet - link
- 15.09 - Please get the frameworks installed by the next class - issue
- 15.09 - added project rules and examples (see "course stuff" below)
- 15.09 - week0 assignment published (see "syllabus")
- week0 Recap
- Lecture: Linear models, stochastic optimization, regularization
- Seminar: Linear classification, sgd, modifications
- HW due: 28.09.16, 23.59.
- Please get bleeding edge theano+lasagne installed for the next seminar.
- week1 Getting deeper
- Lecture: Neural networks 101
- Seminar: theano, symbolic graphs and basic neural networks
- HW due: 3.10.16 23.59
- week2 Deep learning for computer vision 101
- Lecture: Convolutional neural networks
- Seminar: lasagne and CIFAR
- HW due: 9.10.16 23.59 on first submission.
- week3 Deep learning for natural language processing 101
- Lecture: NLP problems and applications, bag of words, word embeddings, word2vec, text convolution.
- Seminar: Text convolutions for Avito content filtering task
- HW due: 16.10.16 23.59 on first submission.
- week4 Recurrent neural networks for sequences
- Lecture: Simple RNN. Why BPTT isn't worth 4 letters. GRU/LSTM. Language modelling. Optimized softmax. Time series applications.
- Seminar: Generating laws for pitiful humans with mighty RNNs.
- HW due: 28.10.16 23.59 on first submission.
- week5 Recurrent neural networks II
- Lecture: Batchnorm and dropout for RNN; Seq2seq: machine translation, conversation models, speech recognition and more. Attention. Long term memory architectures.
- Seminar: a toy machine translation task
- to be anounced
- [Skip week]
- week6 Fine-tuning with neural networks
- Lecture: Large CV datasets, model zoo, reusing pre-trained networks, fine-tuning, "knowledge transfer", soft-targets
- Seminar: Cats Vs Dogs Vs Very Deep Networks
- HW due 17.11.16 23.59
- week7 Advanced computer vision
- Lecture: Representations within convnets, fully-convolutional networks, bounding box regression, maxout, etc.
- Seminar: Image captioning by Arseniy Ashukha
- HW due 24.11.16 23.59
- week8: Generative models for computer vision
- Lecture: Autoencoders, Generative Adversarial Networks
- Seminar: Art Style Transfer with deep learning (Dmitry Ulyanov)
- HW due 4.12.16 23.59
- week8: Deep learning for sound processing
- Lecture: case study: music recommendation with deep learning
- Seminar: Music clustering & content-based recommentation with convolutional nets
- HW due 11.12.16 23.59
- week9: Basic reinforcement learning
- Lecture: Introduction to reinforcement learning
- Seminar: one algorithm to navigate in a maze, play pacman and control robots.
- HW due 11.12.16 23.59
- week10: Deep reinforcement learning
- Lecture: approximate reinforcement learning with deep neural networks (problems and solutions)
- Seminar: Playing Atari/Doom with deep reinforcement learning
- HW due 18.12.16 23.59 first submission
- week12: Bayesian deep learning
- Lecture: Basics of bayesian approach to probabilities
- Bonus lecture: Variational autoencoders (Mikhail Khalman)
- HW due 26.12.16 16.00 hard
- One rule to rule them all
- Project rules
- Project examples
- Grading system
- Grading table
- Reducing lateness penalty
- Feedback form (anonymous)
Course materials and teaching performed by
- Fedor Ratnikov - lectures, seminars, hw checkups
- Oleg Vasilev - seminars, hw checkups, technical issue resolution
- Arseniy Ashukha - image captioning, sound processing, week7&9 lectures
- Dmitry Ulyanov - generative models, week8 lecture, week12 homework assignment
- Mikhail Khalman - variational autoencoders, lecture 12
- Vadim Lebedev - week0 & week6 homeworks
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent