Machine Learning

Documents

Slides

Einführung (Version 1.1)
Kategorisierung (Version 1.1)
Ähnlichkeitsmaße und Vorverarbeitung (Version 1.0)
Bayessche Entscheidungstheorie (Version 1.2)
Naive Bayes Klassifizierer (Version 1.0)
Entscheidungsbäume (Version 1.0)
Lineare Diskriminanz (Version 1.1)
Neuronale Netze Teil 1 (Version 1.2)
Neuronale Netze Teil 2: MLP (Version 1.0)
Programmierung Neuronaler Netze mit Pybrain(Version 1.0)
Support Vector Machines (Version 1.1)
Unüberwachtes Lernen: Clustering (Version 1.1)
Reinforcement Learning (Version 1.0)

The slides for SS 10 will be uploaded immediately before the corresponding lecture.

Python

Python Tutorial (Maucher) (Version 1.0)

Exercises

Exercise 1: Similarity
Task 1 description
kundenkarte.txt
gekaufteProdukte.txt
nachrichtentexte.txt

Exercise 2: Product Similarity and Recommendations
Task 2 description

Exercise 3: Parametric Classification and Regression
Task 3 description
Trainingdata AutoKunden.txt
Trainingdata HeartRate.txt
Trainingdata laktat.txt

Exercise 4: Naive Bayes Classification
Task 4 description
Training Data Master Decision
Feed-Parser Example
Universal Feed Parser

Exercise 5: Decision Tree
Task 5 description
Trainingdata Notebooks
Trainingdata Notebooks (reduced in .arff)

Exercise 6: Perceptron Learning Rule
Task 6 description
Python GUI Template

Exercise 7: MLP for classification and regression
Task 7 description
Template for regression program

Exercise 8: Support Vector Machines
Task 8 description

Exercise 9: Unsupervised Learning: Colorspace quantisation
Task 9 description

Exercise Collection

Exercise Collection and Solutions

Additional Reading

General Information on the Lecture, Term SS 10:

EDV Nr.	38355
Date:	Mo 14.15h-17.30h
Room:	136
SWS/ECTS:	4/6

Contents

The science of Machine Learning seeks to answer the question

How can we build computer systems that automatically improve with experience, and what are the fundamental laws that govern all learning processes?

In contrast to conventional computer systems, adaptive systems that integrate Machine Learning algorithms do not only process data, but try to extract knowledge from the available data and apply this knowlege for self-correction and -optimisation.
Machine Learning can be considered as an intersection of Computer Science and Statistics. Whereas Computer Science has focused primarily on how to manually program computers, Machine Learning focuses on the question of how to get computers to program themselves. Whereas Statistics already deals with the problem of how to infer knowledge out of a given set of data and some modelling assumptions, Machine Learning incorporates additional questions about what computational architectures and algorithms can be used to most effectively capture, store, index, retrieve and merge these data. Machine Learning also applies results and models from neuroscience and psychology
Machine Learning is an essential part of Artificial Intelligence. Data-/Web-Mining and Pattern Recognition can be considered as Machine Learning applications. A more concrete but not complete list of applications is:

Robotics
Marketbasket analysis, customer analysis and recommender systems
Object Recognition (including character- speech- and face recognition) and Computer Vision
Search Engines
Sentiment Analysis, Opinion Mining
Gaming
Speech- and ImageCompression
...

The lecture aims to provide the necessary theory as well as insights into the most important applications of Machine Learning. In the excercises students learn to program algorithms and small applications. Python is applied for this programming since it is easy to learn and provides a bunch of helpful packages.

The lecture is structured as follows

Introduction, Overview and Categorisation
Relations between Machine Learning, Artificial Intelligence and Data Mining
Bayesion Learning Theory
Decision Trees
Neuronal Networks
Support Vector Machines
Learning Association Rules
Cluster Analysis
Feature Selection and Feature Extraction
PCA-based Face-Regognition
Reinforcement Learning

An excerpt of applications, which will be programmed in the exercises:

Recommender Systems
Document Classification
Predictions based on linear- and non-linear regression
Character recognition
Face recognition
Colorspace transformation
...

In the first exercise an introduction into the basic concepts of Python will be provided.

Announcements

15.03.2010: First lesson in this term. Delayed start at 16.00h due to the newcomers intro.