| 2-INF-150: Strojové učenie / Machine Learning Fall 2022 Handouts |
Contact | Basic information | Homework assigments | Handouts | Previous semesters
This page shows preliminary schedule of classes and servers as a repository of materials relevant to class. The schedule will be update regularly after lectures.
Recommended literature:
In the schedule, we list the chapters most relevant to the material covered in class. Presentation of the material in lectures usually differs from the books. The book chapters should serve mainly as an additional materials for self study.
Additional materials:
| Week 19.-23.9.2022 | |
|
Administration. Introduction. Supervised learning / regression. Linear regression and it's variants. Literature: GBC:2.1-2.4 or B:C; GBC:5.1 or B:3.1 or HTF:3.1-3.2 |
|
| Slides and notes: | Supporting materials: |
|---|---|
|
video intro: [ linka ] video regression: [ linka ] slides 1: [ PDF, 275 Kb ] slides 2: [ PDF, 342 Kb ] |
stanford (chapters 1, 2, 4): [ linka ] |
| Week 26.9.-30.9.2022 | |
|
Tutorials 1: collabora/jupyter notebooks, numpy Theory of learning, overfit, underfit, bias variance. Literature: GBC:4.3,5.9; B:3.1; |
|
| Slides and notes: | Supporting materials: |
|---|---|
|
Tutorials 1: numpy: [ linka ] Learning theory notes: [ PDF, 362 Kb ] video lecture: [ linka ] |
stanford (chapter 4): [ linka ] stanford (chapter 1): [ linka ] |
| Week 3.-7.10.2022 | |
| Theory of learning cont. Regularization. Classification. Logistic regression, softmax (maximum entropy) classifier. Neural networks. | |
| Slides and notes: | Supporting materials: |
|---|---|
|
lecture: [ linka ] lecture 2 (until 35 minute, stop at neural networks): [ linka ] |
stanford (chapter 3, 5 and 9.3): [ linka ] stanford (chapter 1): [ linka ] |
| Week 10.-14.10.2022 | |
|
Tutorials 2: regression Support vector machines. |
|
| Slides and notes: | Supporting materials: |
|---|---|
|
SVM lecture: [ linka ] Tutorials 2: Regression: [ linka ] |
|
| Week 17.-21.10.2022 | |
|
Tutorials 3: Neural networks Support vector machines (continued). Kernel trick. |
|
| Slides and notes: | Supporting materials: |
|---|---|
|
Tutorials 3: neural networks: [ linka ] lecture1: [ linka ] lecture2: [ linka ] |
SVM Stanford: [ linka ] |
| Week 24.-28.10.2022 | |
| Decision trees. Bagging and boosting. Probabilistic interpreration of regresion and regularization. | |
| Slides and notes: | Supporting materials: |
|---|---|
|
lecture: [ linka ] |
Decision Trees: [ linka ] Decision Trees 2: [ linka ] Gradient boosting: [ linka ] |
| Week 31.10-4.11.2022 | |
|
Tutorials 4: trees and SVMs Theory of ML (PAC learning). Estimating needed number of training samples and expected test error for finite hyphothesis set. Estimation for infinite set of hyphotheses. VC dimension. PAC learning and SVM. |
|
| Slides and notes: | Supporting materials: |
|---|---|
|
Tutorials 4: trees and SVMs: [ linka ] PAC learning rectangle game: [ linka ] PAC learing, VC dimension (until 50 minute): [ linka ] |
PAC - finite hyphotheses: [ PDF, 243 Kb ] axis aligned rectangles / Andres Munoz, NY Univ: [ PDF, 97 Kb ] VC dimension - definition and examples / Yishai Mansour, U Tel Aviv: [ PDF, 109 Kb ] VC dimension - PAC estimates / Yishai Mansour, U Tel Aviv: [ PDF, 173 Kb ] PAC estimates for SVM: [ PDF, 295 Kb ] |
| Week 7.11-11.11.2022 | |
|
PAC learning continued Principal component analysis (PCA) |
|
| Slides and notes: | Supporting materials: |
|---|---|
|
Video lecture PCA: [ linka ] |
PCA: [ PDF, 567 Kb ] |
| Week 14.-18.11.2022 | |
| Tutorials 5: PCA | |
| Slides and notes: | Supporting materials: |
|---|---|
|
Tutorials 5: PCA: [ linka ] |
|
| Week 21.-25.11.2022 | |
|
Clustering (k-means, k-medoids, hierarchical clustering) Reinforcement learning |
|
| Slides and notes: | Supporting materials: |
|---|---|
|
Clustering slides: [ PDF, 185 Kb ] Clustering lecture (from minute 50): [ linka ] Reinforcement learning notes (contain more links at the end): [ linka ] |
|