Methodologies for Intelligent Systems
Course Aim & Organization
The objective of this course is to give an advanced presentation of the techniques most used in artificial intelligence and machine learning for pattern recognition, knowledge discovery, and data analysis/modeling.
The course is composed by a blending of lectures and exercises by the course teacher and some teaching assistants.
- Matteo Matteucci: course teacher
- Davide Eynard: teaching assistant on clustering
- Rossella Blatt: teaching assistant on feature selection/projection
Course Program and Schedule
These techniques are presented from a theoretical (i.e., statistics and information theory) and practical perspective through the descriptions of algorithms, their implementation, and applications.The course is composed by a set of selfcontained lectures on specific techniques such as decision trees, decision rules, Bayesian networks, clustering, etc. Supervised and unsupervised learning are discussed in the framework of classification and clustering problems. The course outline is:
- 'Machine Learning and Pattern Classification': in this part of the course the general concepts of Machine Learning and Patter Recognition are introduced with a brief review of statistics and information theory;
- Unsupervised Learning Techniques: the most common approaches to unsupervised learning are described mostly focusing on clustering techniques, rule induction, Bayesian networks and density estimators using mixure models;
- Supervised Learning Techniques: in this part of the course the most common techniques for Supervised Learning are described: decision trees, decision rules, Bayesian classifiers, hidden markov models, lazy learners, etc.
- Feature Selection and Reduction: techniques for data rediction and feature selection will be presented with theory and applications
- Model Validation and Selection: model validation and selection are orthogonal issues to previous technique; during the course the fundamentals are described and discussed (e.g., AIC, BIC, cross-validation, etc. ).
A detailed schedule of the course can be found here; topics are just indicative while days and teachers are corret "up to some last minute change".
Lecture Slides on Machine Learning and Pattern Recognition
- Lecture 1: Introduction to Machine Learning
- Lecture 2: Probability for Dataminers
- Lecture 3: Decision Trees
- Lecture 4: Decision Rules
- Lecture 5: Bayesian Classifiers
- Lecture 6: Bayesian Networks
- Lecture 7: Markov Chains and Hidden Markov Models
Lecture Slides on Clustering
- Clustering Lecture 1: Introduction (03/04/2008)
- Clustering Lecture 2: K-Means and Hierarchical (03/04/2008)
- Clustering Lecture 3: Fuzzy, Gaussians, and SOM (03/04/2008)
- Clustering Lecture 4: Vector Spacec Model and PDDP
- Clustering Lecture 5: DBSCAN and Jarvis Patrick
- Clustering Lecture 6: Evaluation measures
Lecture Slides on Dimensionality Reduction and Feature Selection
- Dimensionality Reduction Lecture 1: Dimensionality reduction Intro
- Dimensionality Reduction Lecture 2: Feature extraction, PCA and LDA
- Dimensionality Reduction Lecture 3: Feature selection
- Genetic Algorithms: a rather comprehensive tutorial
- Algorithm Evaluation: from cross-validation to confidence intervals
A few additional lecture notes
- Fundamental Problems for HMM: a document to introduce Hidden Markov Models and the three fundamental questions about them.
- An exercise on modeling and reasoning with Bayesian Networks.
The homework, although not mandatory, counts for the 30% of the course grade (i.e., if you do not turn it in you loose 30% of the final grade). You have 15 days to turn it in to the teacher. This year the homework is due by the 3rd of July!
- Homework for the academic year 2008/2009
Past years course homework; you can use them to make some practice and prepare this year homework ;-)
- Homework for the academic year 2007/2008
- Homework for the academic year 2006/2007
- Homework for the academic year 2005/2006