460-4053/01 – Biologically Inspired Computing (BIV)
| Gurantor department | Department of Computer Science | Credits | 4 |
| Subject guarantor | prof. Ing. Ivan Zelinka, Ph.D. | Subject version guarantor | prof. Ing. Ivan Zelinka, Ph.D. |
| Study level | undergraduate or graduate | | |
| | Study language | Czech |
| Year of introduction | 2010/2011 | Year of cancellation | 2015/2016 |
| Intended for the faculties | FEI | Intended for study types | Follow-up Master |
Subject aims expressed by acquired skills and competences
The goal is to introduce the students with modern methods of calculation derived from evolutionary and biological processes (evolutionary algorithms, cellular automata etc.). Student will gain an overview of modern computer-based procedures principles of observation of biological processes and dynamics. Upon successful completion of graduate course will be able to apply the methods discussed in the course to real problems.
Teaching methods
Lectures
Summary
The course will discuss a wider range of evolutionary computation. They
mentioned as historically classic techniques and modern algorithms. There will also be discussed at the introductory level, cellular automata, artificial life, neural networks, evolutionary hardware, DNA computing, etc. Great emphasis will be placed on the practical side of things - the ability to most discussed methods applied to practical examples. Students should have the comprehensive knowledge of the course of the above areas, including the possibility its use. The course includes laboratory exercises in which students will practice
how to program the selected algorithms and their application to solving practical
problems.
Compulsory literature:
Recommended literature:
Additional study materials
Way of continuous check of knowledge in the course of semester
Kontrola je založena na vypracovávání protokolů předmětu, pomocí kterých student prokazuje nejen pochopení informací z přednášek, ale i schopnost jejich implementace v daném programovém prostředí. K získání zápočtu je nutno odevzdat cvičícímu všechny požadované protokoly a mít alespoň 80% fyzické účasti na laboratořích. Zápočet je podmínkou NUTNOU k připuštění ke zkoušce.
U studentů kombinovaného studia jsou laboratoře nahrazeny vypracováním zadaných protokolů.
E-learning
Other requirements
It is required the ability to create programs in arbitrary programming language.
Additional requirements are not defined.
Prerequisities
Subject has no prerequisities.
Co-requisities
Subject has no co-requisities.
Subject syllabus:
Lectures:
1.Evolutionary algorithms 1. The current state of the field softcomputing, fuzzy logic, neural
networks, evolutionary computing (EVT). Classification of evolutionary computation,
historical facts, current trends in EVT. The central dogma, according to Darwin, and EVT
Mendel.
2. Evolutionary Algorithms 2. No Free Lunch Theorem. Computational complexity and physical limitations
algorithms. Multipurpose optimization and Pareto set.
3. Evolutionary algorithms 3. Restrictions placed on the utility function and individual parameters.
Penalties and its impact on the geometry of the objective function. Working with real, integer and
discrete values of individual parameters. Genetic algorithms. GA terminology. Principle
activities, Hybrid GA, messy GA, parallel GA, migration and diffusion models.
4. Evolutionary algorithms 4. Evolutionary Strategy. No-man (1 +1)-EC.Multi-EC (μ
+ Λ)-ES and (μ, λ)-ES. Multi-EC (μ + λ)-ES and (μ, λ)-ES. Adaptive EC.Particle swarm
(Particle Swarm). Search suspended (Scatter Search). Ant colony optimization (ant
Colony Optimization).
5. Evoluční algorithms 5. SOMA: Self-Organizing Migrating Algorithm principle of operation
and strategies used by the algorithm: ATO, ATR ATAA and ATA. Differential evolution principle
activities and used versions: DE/best/1/exp, DE/rand/1/exp, DE/rand-to-best/1/exp, DE/best/2 /
exp DE/rand/2/exp, DE/best/1/bin, DE/rand/1/bin, DE/rand-to-best/1/bin, DE/best/2/bin,
DE/rand/2/bin. SOMA, DE and permutation test problems.
6. Evolutionary algorithms 6. Techniques of Genetic Programming: Genetic Programming,
grammatical evolution. Alternatives: analytical programming, Probabilistic Incremental
Program Evolution - PIPE, Gene Expression Programming, Programming Multiexpression
and more.
7. Evolutionary Hardware (EH). Inspiration in biology. Computing devices.Reconfigurable
equipment. Evolutionary design and digital circuits. EH and cellular automata. Polymorphous
electronics.
8. Cellular Automata (BA) and complex systems. Introduction, Formalism BA
Dynamics and classification according to Wolfram's cellular automata, Conway's Game of Life,
using BA modeling.
9. Artificial life. Basic definitions and existing systems and models. Tierra, biomorf, Sbeat,
Sbart, Eden, Galapagos ... Self-reproducing automata according to Turing and von
Neumann. Langton's loop, computer viruses and artificial life. Artificial Life and edge
chaos (according to Kaufmann)
10. Neural Networks (ANN). History and basic principles of NS. The training set and its use
NS. The basic types of networks and their applications to different types of problems.
11. Fractal geometry. History, definition of fractal, basic types of algorithms that generate
fractals. Fractal dimension, interpolation and compression. Developmental systems and artificial life. L systems,
turtle graphics, parametric L-systems, L-systems from the perspective of fractal geometry.
12. Immunological systems (IS). The principle of the IS, the IS limits, algorithms implementing IS imunotronics.
13. Swarm Intelligence (SI). Basic concepts and definitions, representative algorithms SI -
Particle Swarm, scatter search, ant colony optimization, swarm robotic, artificial evolution
complex systems.
14. DNA computing. DNA computing as part of bioinformatics, DNA and binary
representation according to Adlemann. Watson Crickův machine. Mathematical modeling operations on DNA.
Laboratories (for PC classrooms):
The seminar will focus on the practical application of the discussed techniques and solutions of selected problem examples.
- Creation of a single basic framework for bio-inspired algorithms on the principles of GUI, 1 week
- Creation of a module for generating population and fitness function, 1 week
- Creation of a module selection techniques for parents (suitable candidates) to create offspring (better solution), 1 week
- Creation of a module for crossover, 1 week
- Creation of a modules of evolutionary algorithms, 4 weeks
- Creation of a modules of symbolic regression, 4 weeks
- Creation of a module with a simple cellular automaton, 1 week
Conditions for subject completion
Conditions for completion are defined only for particular subject version and form of study
Occurrence in study plans
Occurrence in special blocks
Assessment of instruction