![[Lectures - 2020/2021]](hg.php?text=Lectures - 2020/2021)
|
Note: This is work in progress, and not final yet!
Changelog: 2020-09-30 Corrected the link to the PDF of the first lecture. No one complained until now, but my apologies anyway. 2020-10-05 Corrected the link to the video of the first lecture. No one complained until now, but my apologies anyway. Slides for the second lecture added. 2020-10-12 Slides for the third lecture added. 2020-10-27 Added slides for the fourth lecture and a local copy of videos for the third lecture. 2020-11-02 Slides for the fifth lecture added. 2020-12-07 Converted videos of lectures 5-9 uploaded to local storage. 2020-12-08 Slides for the ninth lecture added. Added information about home assignments #3, #4 and #5. 2020-12-14 Added video for image recognition.
|
Lecture date
|
Title
|
1. |
September 21
|
Lecture: Introduction. Linear algebra and Matlab review. Polynomials and Taylor series. Euler's identity. [PDF]
Selected linear algebra resources:
- you own lecture notes from 11LA or a similar course
- free Czech lecture notes by Petr Olšák
- free English textbooks just randomly found and not reviewed:
- for videos, try Khan Academy, MIT OpenCourseWare, and similar, and consult also YouTube
Selected Matlab resources:
- search in webpages of lectures 11MSP, 11MAG (in Czech), or 20SK (in English)
- Czech text Jemný úvod do Matlabu a Simulinku (the initial part about Matlab is enough),
- books by Cleve Moler available from MathWorks web (in English),
- partially outdated Czech book by Pavel Karban Výpočty a simulace v programech Matlab a Simulink (Computer Press, 2006, 224 pp., ISBN: 978-80-251-1448-3),
- educational modules at UNSW a HIT (in English),
- Google "Úvod do Matlabu" or "Introduction to Matlab" and similar keywords,
- the same applies to YouTube, there is a plethora of video tutorials for Matlab and some of them are actually quite good.
Recording of the online session is available on the MS Teams channel. Note that it expires after 20 days. The local copy is here.
|
- |
September 28
|
State holiday.
|
2. |
October 5
|
Lecture: Vector spaces, signals, and images. Series and Fourier Series. Introduction to Discrete Fourier Transform. Aliasing, zero-padding. [PDF]
Computer session: Test of basic linear algebra and Matlab skills. Additional resources: Matlab script, Jupyter/iPython notebook pre-rendered into HTML
Recording of the online session is available on the MS Teams channel. Note that it expires after 20 days. The local copy of the recording in English is here (lecture) and here (computer session) and the recording in Czech is here (lecture) and here (computer session).
|
3. |
October 12
|
Lecture: Discrete Fourier basis. Short-Time Fourier Transform. Windowing and localisation. [PDF]
Computer session: Finding spectra with MATLAB. Aliasing, zero-padding, resampling. Additional resources: Matlab script archive
Recording of the online session is available on the MS Teams channel. Note that it expires after 20 days. The local copy of the recording in English is here (lecture) and here (computer session) and the recording in Czech is here (lecture) and here (computer session).
|
- |
October 19
|
Canceled
|
4. |
October 26
|
Lecture: Short-Time Fourier Transform. Spectrograms. Time- and frequency resolution. [PDF]
Computer session: The spectrum of a non-stationary signal vs. its spectrogram.
Recording of the online session is not available anywhere due to an unknown MS Teams error.
|
5. |
November 2
|
Lecture: Data processing review from 11MAMY: linear and logistic regression, regressor selection, regularization, logistic classification, discriminant analysis, principal component analysis, clustering [PDF]
Recording of the online session is available on the MS Teams channel. Note that it expires after 20 days. The local copy of the recording in English is split into part 1 and part 2 and the same holds for the recording in Czech -- here is part 1 and part 2.
|
6. |
November 9
|
Lecture: Bootstrap. Advanced methods for regressor selection. [PDF]
Computer session: Regression review. Bootstrapping. PCA regression. Data: islr_auto.csv
Recording of the online session is available on the MS Teams channel. Note that it expires after 20 days. The local copy of the recording in English is here (lecture) and here (computer session) and the recording in Czech is here (lecture) and here (computer session).
|
-
|
November 16
|
Dean's leave
|
7. |
November 23
|
Lecture: Regressor selection: Principal components regression, partial least squares. [PDF]
Computer session: Cleaning and processing a CSV-based dataset in Matlab. Principal components regression example. Data: islr_hitters.csv
Recording of the online session is available on the MS Teams channel. Note that it expires after 20 days. The local copy of the recording in English is here (lecture) and here (computer session) and the recording in Czech is here (lecture) and here (computer session).
|
8. |
November 30
|
Computer session: Partial least squares. Comparison between PCR, PLS, and Lasso on ISLR Hitters dataset.
Recording of the online session is available on the MS Teams channel. Note that it expires after 20 days. The local copy of the recording in English is split into part 1 and part 2, the recording in Czech is shorter and therefore in a single part.
|
9. |
December 7
|
Intro: Review of Assignments #3, #4, and #5 [PDF]
Lecture: Introduction to Neural Networks [PDF]
Recording of the online session is available on the MS Teams channel. Note that it expires after 20 days. The local copy of the recording in English is available in a single part here and the recording in Czech here.
|
10. |
December 14
|
Computer session: Letter recognition [video]. Time series prediction using LSTM ANN [video].
|
11. |
January 4
|
Computer session: Traffic flow forecasting using LSTM ANN [video].
|
|
|
|