https://wikicshse.ru/index.php?action=history&feed=atom&title=Dse_2023-24
Dse 2023-24 - История изменений
2026-06-06T11:52:00Z
История изменений этой страницы в вики
MediaWiki 1.45.3
https://wikicshse.ru/index.php?title=Dse_2023-24&diff=256&oldid=prev
imported>Pshuanar: /* Log Book or Tentative Plan */
2023-10-31T13:12:01Z
<p><span class="autocomment">Log Book or Tentative Plan</span></p>
<p><b>Новая страница</b></p><div>== General course info ==<br />
<br />
16 lectures plus 16 classes<br />
<br />
Fall grade = 0.2 Small HAs + 0.2 Group project + 0.3 Midterm + 0.3 Final<br />
<br />
Each small HA consists of approximately 4 or 5 problems. Group project may be written by a group of 1-3 students. Midterm and Final are offline and hand written. <br />
<br />
Lecturer: Boris Demeshev<br />
<br />
Class teachers: [https://www.hse.ru/org/persons/190922066 Yana Khassan], Shuana Pirbudagova<br />
<br />
[https://t.me/+wieDS2ZjAvVjYTJi tg group], <br />
<br />
Lectures: Monday, 18:10 - 19:30 Moscow time, [https://zoom.us/j/8126338383 zoom]<br />
<br />
Classes: <br />
* Thursday, 13:00 - 14:20 Moscow time, D208, Shuana<br />
<br />
Github repository of the class: [https://github.com/Shuaynat/DSE-23-24/tree/main]<br />
<br />
==Log Book or Tentative Plan ==<br />
<br />
[https://www.youtube.com/playlist?list=PLyjahhN4Wdd9rpesSCbxg5YdtF_cJ1SZC 🔗Lecture playlist]<br />
<br />
[https://www.youtube.com/playlist?list=PLyjahhN4Wdd99xaZOCZ99hNcLiM66QR9E 🔗Consultations playlist]<br />
<br />
'''Week 1. 2023-09-04''': Entropy, [https://github.com/Shuaynat/DSE-23-24/raw/main/03-lectures/Dse2023-L01.pdf pdf]<br />
<br />
Guessing game, conditional entropy, joint entropy.<br />
<br />
Class: data manipulation, data vizualization<br />
<br />
More:<br />
<br />
Cristopher Olah, Visual Information Theory <br />
https://colah.github.io/posts/2015-09-Visual-Information/<br />
<br />
Grand Sanderson, Solving Wordle using information theory<br />
https://www.youtube.com/watch?v=v68zYyaEmEA<br />
<br />
конспект аналогичной лекции на фкн:<br />
https://exuberant-arthropod-be8.notion.site/1-02-09-5e107ea1c4054594b8f37d955db8a2b0<br />
<br />
'''Week 2.''': Kelly criterion, [https://github.com/Shuaynat/DSE-23-24/raw/main/03-lectures/Dse2023-L02.pdf pdf]<br />
<br />
How to calculate expected values using cross-entropy ideas, H(X) - H(X|Q) as long term interest rate. <br />
<br />
More:<br />
<br />
[https://en.wikipedia.org/wiki/Kelly_criterion Kelly criterion]<br />
<br />
Class: group by, reshape and join<br />
<br />
'''Week 3.''': Trees, [https://github.com/Shuaynat/DSE-23-24/raw/main/03-lectures/Dse2023-L03.pdf pdf]<br />
<br />
Class: Trees (regression + classification) + tree visualization<br />
<br />
More:<br />
<br />
[http://www.r2d3.us/visual-intro-to-machine-learning-part-1/ tree visualization by r2d3]<br />
<br />
[https://en.wikipedia.org/wiki/Receiver_operating_characteristic accuracy, recall, roc and all of that]<br />
<br />
'''Week 4.''' Random forest [https://github.com/Shuaynat/DSE-23-24/raw/main/03-lectures/Dse2023-L04.pdf pdf]<br />
<br />
More: <br />
<br />
[http://www.r2d3.us/visual-intro-to-machine-learning-part-2/ bias-variance trade-off visualization for trees]<br />
<br />
[https://arxiv.org/pdf/1411.5279.pdf Tim Hesterberg, What teachers should know about bootstrap?] Very well written text, for permutation test see sections 2.1 and 7. <br />
<br />
Class: Random forest, cross-validation in sklearn, feature importance, <br />
<br />
'''Week 5.''': Gradient boosting, Data splitting strategies<br />
<br />
Class: XGBoost vs LightGBM, Dummy variables, categorical variables and Catboost <br />
<br />
'''Week 6.''': Naive bootstrap, t-stat bootstrap, permutation tests<br />
<br />
Class: Hypothesis testing <br />
<br />
More:<br />
<br />
https://arch.readthedocs.io/en/latest/bootstrap/bootstrap.html<br />
<br />
'''Week 7.''': Matrices in regression<br />
<br />
Class: (by hand) Differential in matrix form, derivation of formulas for beta.<br />
<br />
'''Here will be <del>dragons</del> midterm!'''<br />
<br />
<br />
'''Week 8.''': SVD = PCA<br />
<br />
Class: (by hand) Covariance matrices, <br />
<br />
<br />
'''Week 9.''': James Stein paradox<br />
<br />
Class: Matrices in numpy, PCA in sklearn, SVD<br />
<br />
'''Week 10.''': L1, L2 regularization<br />
<br />
Class: Regression in sklearn, different type of regularisation<br />
<br />
'''Week 11.''': Log regression + L1/L2<br />
<br />
Class: Log regression (sklearn/statsmodels) + L1/L2<br />
<br />
'''Week 12.''': Hierarchical clustering + k-means<br />
<br />
Class: Hierarchical clustering + k-means<br />
<br />
'''Week 13.''': ETS (Exponential Smoothing)<br />
<br />
Class: Plotting time series, ETS (sktime)<br />
<br />
More: <br />
<br />
https://www.sktime.net/en/stable/examples/01_forecasting.html<br />
<br />
'''Week 14.''': Bayesian approach<br />
<br />
Class: TS forecasting with grad boosting <br />
<br />
'''Week 15.''': Mention of MCMC + DLT<br />
<br />
Class: DLT in python<br />
<br />
More: <br />
<br />
Mcmc visualization: https://chi-feng.github.io/mcmc-demo/app.html?algorithm=SVGD&target=banana&delay=0 <br />
<br />
https://www.uber.com/blog/orbit/<br />
<br />
'''Week 16.''': QA <br />
<br />
Class: QA<br />
<br />
Here will be <del>dragons</del> final!</div>
imported>Pshuanar