A general introduction to statistical machine learning
Presenters
Nic Schraudolph
Simon Guenter
Jin Yu
Tiberio Caetano
SVN Vishwanathan
Time and Place
Lectures: Mon & Wed 10-12
Tutorials: Fri 10-12
room A207, RSISE
Tutorials are optional and will be held on an as-needed basis.
Assessment
Only a pass or fail mark will be awarded. To pass the course, students must gain a pass mark on at least 3 out of at least 4 offered assignments. Please hand in assignments to Jin Yu.
Textbook
There is no textbook per se, but if you want to delve deeper into this topic, we recommend:
Hastie, Tibshirani, and Friedman:
The Elements of Statistical Learning
(Data Mining, Inference, and Prediction)
Springer Verlag, 2001
Syllabus
| Nr | Date | Presenter | Content |
|---|---|---|---|
| 1 | 26.4. | nic | Bayesian Inference and Maximum Likelihood Modeling |
| A1 | due 1.5. | Assignment 1: Ovarian Cancer Screening | |
| 2 | 1.5. | nic | Density Estimation and Mixture Models and the EM Algorithm (background reading: EM tutorial.) |
| A2 | due 8.5. | Assignment 2: Density Estimation (on the Fisher Iris Data) | |
| 3 | 3.5. | simon | Regression: Least-squares regression, linear vs. non-linear models, basis functions, gradient descent (Slides3) |
| T | 5.5. | jin | Implementation of Density Estimation Methods |
| 4 | 8.5. | simon | Classification: Classifiers (k-Nearest Neighbor, Centroid, Decision Trees,linear,LDA...), Classification using regression, probability distribution (Slides4) |
| A3 | due 22.5. | Assignment 3: Regression Classifier. (on the Fisher Iris Data) | |
| 5 | 10.5. | nic | Introduction to Neural Networks (Lecture 1, and the beginning of Lecture 2) (Slides5, 6.5MB) |
| T | 12.5. | jin | BP algorithm, How to implement a regression classifier |
| 6 | 15.5. | nic | Gradient Methods (first 2 lectures) |
| 7 | 17.5. | simon | Overfitting and Validation Procedures (e.g. Cross-validation) (Slides7) |
| A4 | due 5.6 | Using 5-fold cross validation to find the optimal K value for K-NN classifier (on the Fisher Iris Data) | |
| 8 | 22.5. | simon | Unsupervised Learning: Clustering algorithms (e.g. k-means), PCA / ICA, Novelty detection (Slides8) |
| 9 | 24.5. | vishy | Exponential Families and Kernel Methods (SlidesExp) |
| 10 | 26.5. | vishy | Exponential Families and Kernel Methods |
| S | 29.5. 11am | nic | Seminar: Accelerating Stochastic Gradient Descent (John Dedman building, room GD 35) |
| 11 | 2.6. | tiberio | Graphical Models (SlidesGM1) |
| 12 | 5.6. | tiberio | Graphical Models (SlidesGM2) (SlidesGM3) (SlidesGM4) |
Last modified 2006-06-11 08:42 AM