University of Birmingham > Talks@bham > Artificial Intelligence and Natural Computation seminars > Level-based analysis of stochastic population models

Level-based analysis of stochastic population models

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Hector Basevi.

Speaker’s website: http://www.cs.bham.ac.uk/~lehrepk/

Abstract: Many processes can be described abstractly as drawing a fixed number of samples from a probability distribution, then adjusting the probability distribution based on information obtained from the samples, and repeating. Examples of such processes include evolutionary algorithms, active learning algorithms, and some economic models. When designing such a process, it is often essential to understanding how the parameters of the process influence its hitting time, i.e., the time until a given target state is reached.

This talk describes the level-based theorem, a technique which provides upper bounds on the expected hitting times of such processes. We first illustrate the method on a simple toy example, then give an overview of applications in evolutionary computation, particularly for noisy optimisation, self-adaptation of mutation rates, and in estimation of distribution algorithms. We finish the talk by discussing some open problems.

This talk is part of the Artificial Intelligence and Natural Computation seminars series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

Talks@bham, University of Birmingham. Contact Us | Help and Documentation | Privacy and Publicity.
talks@bham is based on talks.cam from the University of Cambridge.