• View
• Edit
• History
• Print

E As

GramInd.pdf Δ

General

• Introduction to Evolutionary Computing by A.E. Eiben and J.E. Smith, (look at the available second chapter and at the slides)

Featured links

• Representations for Evolutionary Algorithms Franz Rothlauf (tutorial slides for the book Representations for Genetic and Evolutionary Algorithms)
• Global Optimization Techniques (Arnold Neumaier)
• The Genetic Programming Bibliography by W. B. Langdon
• Illinois Genetic Algorithms Laboratory

Propozycje referatów

1. Learning classifier systems (trochę o uczeniu ze wzmocnieniem “reinforcement learning”)
2. ewolucja neuronalna: NEAT vs. rekurencyjne sieci Schmidhubera (moĹźe teĹź o kodowaniu Gruau)
3. ewolucja kontrolerów samochodów wyścigowych (powiązane z powyĹźszymi)
4. programowanie genetyczne: MOSES (ze wzmianką o OpenCogu) i algorytm PLEASURE
5. sztuczne systemy immunologiczne
6. zagadnienia reprezentacji, “Forma Analysis of Genetic Algorithms”

From the classroom

1. Some ideas on Evolutionary Algorithms Δ
   • Learning Classifier Systems for Hyperspectral Images Processing
2. Survey of Global Optimization Δ
   • Global Optimization by Arnold Neumaier
     • Global optimization and constraint satisfaction slides
     • Complete Search in Continuous Global Optimization and Constraint Satisfaction
   • Black Box Optimization with Data Analysis by Kevin Kofler
3. Multistart Simulated Annealing and Graph Coloring
   • Graph Coloring Instances
   • Zadanie1
4. Compact Genetic Algorithm Δ
   • The compact genetic algorithm by Fernando G. Lobo, Georges R. Harik, David E. Goldberg
   • Linkage learning via probabilistic modeling in the ECGA by Georges Harik
   • Zadanie2 (experiment with cross-over and selective pressure on k-deceptive OneMax)
   • Zadanie3 (experiment with gene fixation under binary coding of continuous fitness goal)
   • Projekt1? (evolve OpenSteer agents)
5. Genetic Programming Δ
   • A Field Guide to Genetic Programming by Riccardo Poli
   • Projekt2? (evolve physical constructions, using OpenDynamicsEngine)
6. [Graph coloring results…]
7. Learning Classifier Systems Δ

Some Evolutionary Computation libraries

• wEvo (Java)
• Open BEAGLE (C++)
• ECJ (Java)
• GAUL (C, S-Lang)
• Evolvica (Java)
• genetic (Python) (poor)
• AI-ALife-HOWTO list of libraries
• GEATbx (Matlab) (old)
• DREAM (Distributed Resource Evolutionary Algorithm Machine) parallelization framework

Test Functions

• http://www.geatbx.com/docu/fcnindex-01.html (unconstrained continuous, with graphs)
• http://www-optima.amp.i.kyoto-u.ac.jp/member/student/hedar/Hedar_files/TestGO.htm (unconstrained and constrained continuous, with Matlab code)
• Collection of problem generators (e.g. bitstring multimodal targets, graph coloring, matrix epistasis, constraint satisfaction)

Artificial Life

• Tierra
• Framsticks
• Polyworld

Matlab implementations

Ask me for adapted Octave versions if the Matlab links below don’t work on your Octave.

Some of my Matlab / Octave scripts

matlab.zip Δ

Differential Evolution

• Differential Evolution (on Rainer Storn pages)
• Matlab code from the book Differential Evolution - A Practical Approach to Global Optimization

NEAT

• NEAT Home Page
• Matlab NEAT

eCGA

• eCGAmatlab.zip Δ
• ECGA.pdf Δ

Other methods of optimization and classes of optimization problems

1. Continuous optimization, mostly
   1. Linear programming
   2. Nonlinear programming
      1. Convex programming
      2. Conjugate gradient method
      3. Lagrange multipliers for constrained problems
   3. Global optimization
2. Combinatorial optimization (discrete optimization, mostly)
   1. Integer programming
   2. Hill climbing
   3. Branch and bound
   4. Simulated annealing
   5. Tabu search
   6. Beam search
   7. Ant colony optimization

Classification of methods

1. Iterative, one-point, local optimization
   • Moves from one solution to a better one
     1. Hill climbing
     2. Gradient descend
2. Monotonically refines a proposition about the search space
   1. Branch and bound
3. Non-monotonically refines a randomized hypothesis about the search space
   1. Simulated annealing
   2. Evolutionary algorithms

Other evolutionary & optimization links

• Theory of simulated annealing, presentation by Nguyen Thuy
• Nonlinear Programming from Applied Mathematical Programming by Bradley, Hax, and Magnanti (Addison-Wesley, 1977)
• Forma Analysis of Genetic Algorithms, Annotated bibliography, Nicholas J Radcliffe
• Learning Classifier Systems: A Survey, Olivier Sigaud, Stewart W. Wilson, 2007
• Branch and Bound Algorithms - Principles and Examples., Jens Clausen