This paper describes the so-called Differential Ant-Stigmergy Algorithm (DASA), which is an extension of the Ant-Colony Optimization for continuous domain. A performance study of the DASA on a benchmark of real-parameter optimization problems is presented. The DASA is compared with a number of evolutionary optimization algorithms including covariance matrix adaptation evolutionary strategy, dif...

This article introduces a model-based reinforcement learning (RL) approach for continuous state and action spaces. While most RL methods try to find closed-form policies, the approach taken here employs numerical on-line optimization of control action sequences. First, a general method for reformulating RL problems as optimization tasks is provided. Subsequently, Particle Swarm Optimization (PS...

The paper presents the Mixed-Integer Non-Linear Programming (MINLP) approach to structural optimization. MINLP is a combined discrete/continuous optimization technique, where discrete binary 0-1 variables are defined for optimization of discrete alternatives and continuous variables for optimization of parameters. The MINLP optimization is performed through three steps: i.e. the generation of a...

Optimization problems attempt to find out the minimum or maximum of a certain function (usually referred to as the cost function). The cost function can either be continuous or discrete. Discrete optimization problems arise, when the variables occurring in the optimization function can take only a finite number of discrete values and also subject to constraint conditions. In continuous optimiza...