Berlin 2015 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

SOE: Fachverband Physik sozio-ökonomischer Systeme

SOE 8: Poster

SOE 8.14: Poster

Monday, March 16, 2015, 18:00–20:00, Poster E

Search space approximation for multi-objective optimization problems — •Thorsten Rehberg — University of Regensburg

A parallel optimization strategy based on simulated annealing and a machine learning approach is used for the approximation of cost functions. This approach is exemplary applied to the portfolio selection problem for developing an efficient cost-function. For this purpose stock data from DAX, MDAX and S&P500 is used for training and validating the results.