SC:Soft Computing

From Chrome
Revision as of 00:39, 5 December 2011 by Matteo (Talk | contribs) (Detailed course schedule)

Jump to: navigation, search


The following are last minute news you should be aware of ;-)

 06/10/2011: the Soft Computing course starts today!

Course Aim & Organization

Soft Computing includes technologies (Fuzzy Systems, Neural Networks, Stochastic Algorithms and models) to model complex systems and offers a powerful modeling tool for engineers and in general people needing to model phenomena. Among the application areas, we mention: data analysis, automatic control, modeling of artificial and natural phenomena, modeling of behaviors (e.g., of users and devices), decision support.

The course will introduce rigorously the fundamentals of the different modeling approaches, will put in evidence the application possibilities, by comparing different models, examples and application cases, will introduce design techniques for systems based on these technologies.

Teachers

The course is composed by a blending of lectures and exercises by the course teacher and the teaching assistant:

Course Program

  • What is Soft Computing: fuzzy systems, neural networks, stochastic algorithms and models;
  • Fuzzy models: fuzzy sets, fuzzy logic, fuzzy rules, motivations for fuzzy modeling;
  • Neural networks: basics, supervised and unsuprvised learning, main modelsi, selection and evaluation;
  • Stochastic models: basics, optimization of models, fitness function, model definition, genetic algorithms, reinforcement learning, bayesian networks;
  • Applications: motivations, choices, models, case studies.

Detailed course schedule

A detailed schedule of the course can be found here; topics are just indicative while days and teachers are correct up to some last minute change (they will be notified to you by email).

Date Day Time Room Teacher Topic
06/10/2011 Thursday 14:15 - 16:15 E.G.6 Andrea Bonarini Introduction - Fuzzy sets
10/10/2011 Monday 15:15 - 17:15 S.1.3 Matteo Matteucci Intro to neural networks and Perceptron model
13/10/2011 Thursday 14:15 - 16:15 E.G.6 Andrea Bonarini Fuzzy sets
17/10/2011 Monday 15:15 - 17:15 S.1.3 Matteo Matteucci Hebbian learning, the xor problem, from perceptron to backpropagation
20/10/2011 Thursday 14:15 - 16:15 E.G.6 Andrea Bonarini Fuzzy logic
24/10/2011 Monday 15:15 - 17:15 S.1.3 Matteo Matteucci Feedforward topologies and Backpropagation
27/10/2011 Thursday 14:15 - 16:15 E.G.6 Andrea Bonarini Fuzzy rules - design of fuzzy systems
03/11/2011 Thursday 14:15 - 16:15 --- --- No lecture today
07/11/2011 Monday 15:15 - 17:15 S.1.3 Matteo Matteucci Neural Network exercises
10/11/2011 Thursday 14:15 - 16:15 E.G.6 Andrea Bonarini Fuzzy systems – Applications
14/11/2011 Monday 15:15 - 17:15 S.1.3 Matteo Matteucci Overfitting limitation
17/11/2011 Thursday 14:15 - 16:15 --- --- No lecture today
21/11/2011 Monday 15:15 - 17:15 S.1.3 Matteo Matteucci Bayesian Networks
24/11/2011 Thursday 14:15 - 16:15 E.G.6 Andrea Bonarini Fuzzy systems – Design
28/11/2011 Monday 15:15 - 17:15 S.1.3 Matteo Matteucci Inference in Bayesian Networks
01/12/2011 Thursday 14:15 - 16:15 E.G.6 Andrea Bonarini Fuzzy systems – Design
05/12/2011 Monday 15:15 - 17:15 S.1.3 Matteo Matteucci Bayesian Networks Demo/Exercises
12/12/2011 Monday 15:15 - 17:15 S.1.3 Andrea Bonarini Reinforcement Learning I
15/12/2011 Thursday 14:15 - 16:15 E.G.6 Matteo Matteucci Bayesian Networks Demo/Exercises
19/12/2011 Monday 15:15 - 17:15 S.1.3 Andrea Bonarini Reinfocement Learning – Design
22/12/2011 Thursday 14:15 - 16:15 E.G.6 Andrea Bonarini Reinfocement Learning – Applications
09/01/2011 Monday 15:15 - 17:15 S.1.3 Andrea Bonarini Evolutionary algorithms – Genetic Algorithms
12/01/2011 Thursday 14:15 - 16:15 E.G.6 Andrea Bonarini Genetic Algorithms – Design
16/01/2011 Monday 15:15 - 17:15 S.1.3 Andrea Bonarini Genetic Algorithms – Applications
19/01/2011 Thursday 14:15 - 16:15 E.G.6 Andrea Bonarini Hybrid systems
23/01/2011 Monday 15:15 - 17:15 S.1.3 Andrea Bonarini Closing remarks and exercises

Course Evaluation

The exam is a test done in regular sessions, starting from the end of the lessons. The test is partitioned in two parts, whose evaluation is averaged. For each of them 32 points are available and a minimum of 15 is needed. The average vote must be greater or equal to 18 to pass the exam. Some example from past years are available below. From year 2011 the format of the exam will change a little bit, but the type of questions will analogous. An example of the format for this year will be published later.

This course can be taken as a stand alone course or as a course integrated with Artificial Intelligence. In both cases, the course will be offered at the same time to students taking one or the other format. The exam will be also the same, but, in the case of integrated course, it will have to be passed together with the exam of Artificial Intelligence, as a unique exam, the same day. The same rules apply for the exam of the integrated course, and the marks obtained in SC and AI will be averaged. The difference between the two solutions is that the integrated course can be selected as a unique course in the study plan.

Teaching Material

Right now, the official course website is maintained by Andrea Bonarini at [1]

Teacher Slides

In the following you can find the lecture slides used by the teacher and the teaching assistants during classes:

  • Course introduction: introductory slides of the course with useful information about the grading, and the course logistics.

Books and Papers

  • ...

Useful Links

...

  • ...

Software

...

  • ...

Links to sites of other Soft Computing courses

...

  • ...

Bibliographic resources

...

  • ...

Exam Samples and Results

Past Exams and Sample Questions

These are the text of past exams to give and idea on what to expect during the class exam:

  • ...

Exam Results

From time to time, you can find here results for the Soft Computing exams, please refer to the official course website for up to date news: