Estimation of Distribution Algorithm (EDA) Projects 

at Essex

   Estimation of Distribution Algorithms for optimisation problems maintain a probability model at each generation, new trial solutions are sampled from this model. The probability model is updated based on statistical information extracted from these trail solutions. 

    Some other heuristics also use probability model for guiding the search, for details, please visit the homepage of probability model based heuristics.


People involved:

John Ford, Tim Gosling, Mathias Kern, Hui Li, Riccardo Poli, Jianyong Sun, Edward Tsang and Qingfu Zhang



A Rigorous Investigation into Estimation of Distribution Algorithms (2002-2003)

Funded by EPSRC (GR/R64742/01)


IGR Report: Summary of the project (pdf)


A New Method for Hard Optimisation and Search Problem (2001-2002)

Funded by University of Essex


Selected Publications:

Journal papers:

Q.  Zhang,  "On the Convergence of a Factorized Distribution Algorithm with Truncation Selection",  Complexity, Accepted for publication, 2004. 

Q.  Zhang,  "On Stability of Fixed Points of Limit Models of Univariate Marginal Distribution Algorithm and Factorized Distribution Algorithm (pdf) ",  IEEE Trans. on Evolutionary Computation, Vol. 8,  No.1, 2004

Q. Zhang, H. Muehlenbein, "On the Convergence of a Class of Estimation of Distribution Algorithms (pdf)", IEEE Trans. on Evolutionary Computation, Vol. 8, No. 2, 2004

Q. Zhang, J. Sun, E. Tsang and J. Ford, "Hybrid Estimation of Distribution Algorithm for Global Optimization", Engineering Computations, Vol. 21, no.1, 2004, pp 91-107.


Conference papers:

Q. Zhang, J. Sun, E. P. K. Tsang and J. Ford, "Combination of Guided Local Search and Estimation of Distribution Algorithm for Solving Quadratic Assignment Problem (pdf)",  Proceedings of the Bird of a Feather Workshops, Genetic and Evolutionary Computation Conference, pp.42 - 48, 2003.

H. Li, Q. Zhang,  E. P. K. Tsang and J. Ford, "Hybrid Estimation of Distribution Algorithm for Multiobjective Knapsack Problem", the 4th European Conference on Evolutionary Computation in Combinatorial Optimization, 5-7 April 2004, Coimbra, Portugal.


Applications of EDAs:

bullet Mathias Kern uses PBIL on dynamic scheduling
bullet Tim Gosling uses PBIL for studying automatic bargaining and game theory



 maintained by Qingfu Zhang
 Last updated: 10/02/04.