Интеллектуальный анализ данных (О.Ю. Бахтеев, В.В. Стрижов)/Осень 2022

Материал из MachineLearning.

(Различия между версиями)
Перейти к: навигация, поиск
Текущая версия (05:28, 14 сентября 2022) (править) (отменить)
(Course page, and projects)
 
(22 промежуточные версии не показаны)
Строка 6: Строка 6:
=Intelligent data analysis=
=Intelligent data analysis=
-
This course delivers methods of model selection in machine learning and forecasting. The modelling data are videos, audios, encephalograms, fMRIs and another measurements in natural science. The models are linear, tensor, deep neural networks, and neural ODEs. The practical examples are brain-computer interfaces, weather forecasting and various spatial-time series forecasting. The lab works are organised as paper-with-code reports.
+
This course develops skills of communication. The goal is to deliver your message to wide auditory of professionals. The form of delivery is a short paper. It results several discussions in our team according to the plan below.
==Schedule and grading==
==Schedule and grading==
-
*[Sep] 16, 23, 30
 
-
*[Oct] 7, 14, 21, 28
 
-
*[Nov] 4, 11, 18, 25
 
-
*[Dec] 2, 9
 
-
# Select topic (report)
+
Workflow
 +
# Select topic (report)
# Prepare material (present 5-10 min and discuss)
# Prepare material (present 5-10 min and discuss)
# Make presentation (20 min and questions)
# Make presentation (20 min and questions)
# Write your text (2 pages and discuss)
# Write your text (2 pages and discuss)
# Publish your text (link)
# Publish your text (link)
 +
 +
Calendar
 +
*Sep: 16, 23, 30 select
 +
*Oct: 7, 14, 21, 28 talk
 +
*Nov: 4, 11 talk, 18, 25 text
 +
*Dec: 2 link, 9 fin
 +
 +
Insert your name and direct link to materials. Each column must carry your name.
 +
 +
{|class="wikitable"
 +
|-
 +
! Date
 +
! Select
 +
! Talk
 +
! Text
 +
|-
 +
|16nxt
 +
|
 +
|Islamov, Strijov
 +
|
 +
|-
 +
|23sep
 +
|...
 +
|
 +
|
 +
|-
 +
|30
 +
|...
 +
|
 +
|
 +
|-
 +
|7oct
 +
|x
 +
|...
 +
|
 +
|-
 +
|14
 +
|
 +
|...
 +
|
 +
|-
 +
|21
 +
|
 +
|...
 +
|
 +
|-
 +
|28
 +
|
 +
|...
 +
|
 +
|-
 +
|4nov
 +
|
 +
|...
 +
|
 +
|-
 +
|11
 +
|
 +
|...
 +
|
 +
|-
 +
|18
 +
|
 +
|x
 +
|...
 +
|-
 +
|25
 +
|
 +
|x
 +
|...
 +
|-
 +
|}
==Course page, and projects==
==Course page, and projects==
-
* TODO Course page
+
* TODO [https://intsystems.github.io/ru/course/ Course page]
-
* TODO Projects
+
* Course repository [https://github.com/intsystems/IDA GitHub]
-
The result links
+
The result links '''before 2nd of december'''
-
* Bronstein, M. [Temporal Graph Networks https://medium.com/towards-data-science/temporal-graph-networks-ab8f327f2efe], Medium TDS
+
* Bronstein, M. [https://medium.com/towards-data-science/temporal-graph-networks-ab8f327f2efe Temporal Graph Networks], Medium TDS
-
*
+
* Benj, E. [https://arstechnica.com/information-technology/2022/09/with-stable-diffusion-you-may-never-believe-what-you-see-online-again/ With Stable Diffusion, you may never believe what you see online again
 +
AI image synthesis goes open source, with big implications], Arstechnica
 +
* MIPT/Strijov, V. [https://www.eurekalert.org/news-releases/871622 Chip controlling exoskeleton keeps patients' brains cool], AAAS ([https://phys.org/news/2018-09-linear-equations-impaired-motion.html variant] Phys.org)
==Topics to discuss==
==Topics to discuss==
-
 
+
* Differential alignment of continuous-time (series) videos [2104.13478]
 +
* Taken's theorem and convergent cross-mapping (signals) [or 2208.10981]
 +
* Graph diffusion models with PDE examples (flows, signals,videos) [2106.10934]
 +
* or probabilistic diffusion models [2208.11970]
 +
* Dimensionality reduction on Riemannian manifolds (for videos) [1605.06182]
 +
* Applications of Lagrangian, Hamiltonian and Noetherian neural PDEs [colab Severilov] [or 2208.06120]
 +
*
==Examples and references==
==Examples and references==
* [https://towardsdatascience.com/questions-96667b06af5#dee8 TDS guidelines]
* [https://towardsdatascience.com/questions-96667b06af5#dee8 TDS guidelines]
 +
* [https://nplus1.dev/blog/2022/04/01/samotek N+1 samotek]
 +
* [https://www.datasciencecentral.com/write-for-us/ DSC write]

Текущая версия

Each Saturday 13:10 at the channel m1p.org/go_zoom


Intelligent data analysis

This course develops skills of communication. The goal is to deliver your message to wide auditory of professionals. The form of delivery is a short paper. It results several discussions in our team according to the plan below.

Schedule and grading

Workflow

  1. Select topic (report)
  2. Prepare material (present 5-10 min and discuss)
  3. Make presentation (20 min and questions)
  4. Write your text (2 pages and discuss)
  5. Publish your text (link)

Calendar

  • Sep: 16, 23, 30 select
  • Oct: 7, 14, 21, 28 talk
  • Nov: 4, 11 talk, 18, 25 text
  • Dec: 2 link, 9 fin

Insert your name and direct link to materials. Each column must carry your name.

Date Select Talk Text
16nxt Islamov, Strijov
23sep ...
30 ...
7oct x ...
14 ...
21 ...
28 ...
4nov ...
11 ...
18 x ...
25 x ...

Course page, and projects

The result links before 2nd of december

AI image synthesis goes open source, with big implications], Arstechnica

Topics to discuss

  • Differential alignment of continuous-time (series) videos [2104.13478]
  • Taken's theorem and convergent cross-mapping (signals) [or 2208.10981]
  • Graph diffusion models with PDE examples (flows, signals,videos) [2106.10934]
  • or probabilistic diffusion models [2208.11970]
  • Dimensionality reduction on Riemannian manifolds (for videos) [1605.06182]
  • Applications of Lagrangian, Hamiltonian and Noetherian neural PDEs [colab Severilov] [or 2208.06120]

Examples and references

Личные инструменты