Evolution Strategies (ESs) and Genetic Algorithms (GAs) are compared in a formal way. It is shown, that both algorithms are identical with respect to their major working scheme, but nevertheless they exhibit signiicant diierences with respect to the details of the selection scheme, the amount of the genetic representation and, especially, the self-adaptation of strategy parameters.

Central problems in path planning are obstacle avoidance and path optimisation. The aim of our work was to investigate the efciency of an Evolution Strategy for optimising robot parameters in both respects. We chose a simple model in order to highlight the effect of evolution as an optimisation procedure. In our model the robot has to reach given goals from a given starting position. In our sim...

We present SPOT, an open-source toolbox for the experimental analysis of optimization algorithms. Evolution strategies (ES) have been severely criticized before they started their triumphal procession as optimization algorithms. Schwefel discussed ideas which appeared to be provocative at the first sight. SPOT provides a modern framework to test these ideas. The extreme programming methodology ...

The paper presents a method based on evolution strategies that attempts to optimise the training parameters of a class of on-line, adaptive connectionistbased learning systems called evolving connectionist systems (ECoS). ECoS are systems that evolve their structure and functionality through on-line, adaptive learning from incoming data. The ECoS paradigm is combined here with the paradigm of e...

Starting Position The starting position is the assumption that one has to cope with experimental optimization, i.e., there is no mathematical description or a simulator for the system to be optimized: you are optimizing a real object at hardware level! Since the interrelationships between the variable input parameters and the dependent output behavior are unknown, we encounter a black box situa...

We discuss massively parallel implementation issues of the following heuristic optimization methods: Evolution Strategy, Genetic Algorithms, Harmony Search, and Simulated Annealing. For the first time, we implement these algorithms on the Connex architecture, a recently designed array of 1024 processing elements. We use the Vector-C programming environment, an extension of the C language adapte...

In this paper, on-line training of neural networks is investigated in the context of computerassisted colonoscopic diagnosis. A memory-based adaptation of the learning rate for the on-line Backpropagation is proposed and used to seed an on-line evolution process that applies a Differential Evolution Strategy to (re-)adapt the neural network to modified environmental conditions. Our approach loo...

This paper describes the adaptation of Evolution Strategies (ESs) for simulation based heat exchanger network synthesis. Due to space limitations the presentation is restricted to a brief overview of the application domain, the problems and pitfalls we encountered during the project and hints to solutions. Some of these problems are of general relevance for the application of Evolutionary Algor...

This paper describes the application of Evolutionary Algorithms (EAs) to the optimisation of a simplified supply chain in an integrated production-inventory-distribution system. The performance of four EAs (Genetic Algorithm (GA), Evolutionary Programming (EP), Evolution Strategies (ES) and Differential Evolution (DE)) was evaluated with numerical sumulations. Results were also compared with ot...

The Natural Evolution Strategies (NES) family of search algorithms have been shown to be efficient black-box optimizers, but the most powerful version xNES does not scale to problems with more than a few hundred dimensions. And the scalable variant, SNES, potentially ignores important correlations between parameters. This paper introduces Block Diagonal NES (BD-NES), a variant of NES which uses...