Loading ...
Sorry, an error occurred while loading the content.

6488PhD studentship in genetic programming

Expand Messages
  • Ekart, Aniko
    Jul 10, 2014

      Dear All,


      I have a PhD studentship available from October 2014, including teaching assistantship on a distance learning programme. I’d be grateful if you could pass this on to any interested graduating BSc or MSc students.


      PhD studentship in Genetic Programming


      Genetic programming (GP) is an established area of computer science, with nearly ten thousand scientific articles published since its inception in the 1980s. The question raised by Samuel in the late 1950s: How could computers learn to solve problems without being explicitly programmed? is to some extent answered by genetic programming. GP automatically evolves computational structures, such as mathematical equations or computer programs. Practically, GP performs an evolutionary search in the space of computational structures and selects the one that solves a given task according to specified criteria.

      The goal of this project is to take GP closer to fully automatic programming by engineering a GP system to exploit emergence and produce good solutions to real life problems.

      Genetic programming systems as such are engineered to produce emergent solutions. Through the use of GP, we expect the computer to solve a problem without explicitly telling it what steps to take and in what order, often not knowing ourselves in advance what the solution should be. If we call the externally observable emergent properties of a system external emergence then in addition to this, internal emergence will allow a system to take advantage of the emergent patterns without external intervention. The ability of self-modification is essential for achieving successful internal emergence. The challenge is to find the appropriate mechanisms for perpetual emergence and autonomous response to emergent properties and incorporate them.

      The decision on the actual real life problems will be part of the project, examples include prediction problems in health, logistics and generative art domains.

      The candidate is expected to have a good degree in Computer Science, good programming skills and interest in engineering and exploiting emergence in genetic programing.

      The project will be supervised by Dr Anikó Ekárt and Dr Christopher Buckingham, who are happy to answer further questions by email to a.ekart@....


      For further details please see: