Oliver Kramer Boeken

Evolutionary algorithms are successful biologically inspired meta-heuristics. Their success depends on adequate parameter settings. The question arises: how can evolutionary algorithms learn parameters automatically during the optimization? Evolution strategies gave an answer decades ago: self-adaptation. Their self-adaptive mutation control turned out to be exceptionally successful. But nevertheless self-adaptation has not achieved the attention it deserves. This book introduces various types of self-adaptive parameters for evolutionary computation. Biased mutation for evolution strategies is useful for constrained search spaces. Self-adaptive inversion mutation accelerates the search on combinatorial TSP-like problems. After the analysis of self-adaptive crossover operators the book concentrates on premature convergence of self-adaptive mutation control at the constraint boundary. Besides extensive experiments, statistical tests and some theoretical investigations enrich the analysis of the proposed concepts.

Focusing on the intersection of machine learning and evolution strategies, this book presents a variety of algorithmic hybridizations that enhance optimization processes. It covers techniques such as covariance matrix estimation, meta-modeling, and clustering-based niching. With a strong emphasis on practical application, experiments utilize a (1+1)-ES implemented in Python with scikit-learn, addressing benchmark problems to illustrate key concepts. The text also explores broader research implications, bridging theoretical frameworks with experimental findings.

This book addresses practical black box optimization challenges, presenting heuristics and algorithms using evolutionary strategies in continuous spaces. It covers evolution strategies, parameter control, and heuristic extensions for constrained and multi-objective problems. It introduces adaptive penalty functions, meta-models, and hybrid methods for efficient optimization, supported by experiments and illustrations.

We are what we hear! — This book focuses on listening and understanding music in educational and informal contexts, starting from the fact that listening is the essential basis for all other musical activities and cultural achievements. Listening is the approach to music for everyone. It is accessible for free and it opens up a world, which is as manifold as possible. To foster the ability to concentrate, to develop sensitivity for sounds, to perceive music as a structural progression, as well as to pay attention to sonic details are important aims of music education. Open ears are the prerequisite for meaningful and expressive music-making. In times of a global music horizon and lifelong learning, we are challenged to expand our awareness and to develop an empathetic state of mind: an appreciative attitude towards new auditory impressions, all kinds of musics, and cultural expressions. European Perspectives on Music Education, Volume 6, contains an international panorama on the topic of listening and understanding music. Chapters are written by authors from Austria, Belgium, Germany, the Netherlands, Norway, Sweden, Switzerland, Turkey, the United Kingdom, and the United States of America, guaranteeing a wide spectrum of musical knowledge, viewpoints, and pedagogical approaches.

This book is devoted to a novel approach for dimensionality reduction based on the famous nearest neighbor method that is a powerful classification and regression approach. It starts with an introduction to machine learning concepts and a real-world application from the energy domain. Then, unsupervised nearest neighbors (UNN) is introduced as efficient iterative method for dimensionality reduction. Various UNN models are developed step by step, reaching from a simple iterative strategy for discrete latent spaces to a stochastic kernel-based algorithm for learning submanifolds with independent parameterizations. Extensions that allow the embedding of incomplete and noisy patterns are introduced. Various optimization approaches are compared, from evolutionary to swarm-based heuristics. Experimental comparisons to related methodologies taking into account artificial test data sets and also real-world data demonstrate the behavior of UNN in practical scenarios. The book contains numerous color figures to illustrate the introduced concepts and to highlight the experimental results. 

Wir leben in einer Kulturwende, die als „iconic turn“ bezeichnet wird: Bildmedien dominieren unsere Wahrnehmungswelt. Zugleich droht die Marginalisierung solcher Erfahrungsbereiche, die wie die Musik als Klanggeschehen zunächst jenseits des Sichtbaren angesiedelt sind. Die Sinne gegeneinander auszuspielen, kann allerdings nicht die Antwort darauf sein. Ästhetische Bildung muss vielmehr deutlich machen, wie Hören und Sehen gemeinsam zur Konstruktion einer integralen Erfahrungswelt beitragen. Wir brauchen Auge und Ohr, um die Welt der Musik in der uns möglichen Erlebnistiefe zu durchdringen: Was sehen wir, wenn wir Musik hören? Mit welchen Anschauungen erschließen wir ihre Struktur, ihren Sinn und ihren Weltbezug? Anliegen dieser Veröffentlichung ist es, die Vielfalt der Verknüpfungsmöglichkeiten von Musik mit visuellen Vorstellungen und Bildern aufzuzeigen und damit zur Differenzierung des musikalischen Erlebens und Verstehens beizutragen.

Computational Intelligence (CI) bezeichnet ein Teilgebiet der Künstlichen Intelligenz, das biologische inspirierte Modelle algorithmisch umsetzt. Evolutionäre Algorithmen orientieren sich an der darwinistischen Evolution und suchen mit Hilfe von Crossover, Mutation und Selektion eine optimale Lösung. Die Fuzzy-Logik ermöglicht als unscharfe Logik eine kognitive Modellierung von Wissen und Inferenzprozessen. Neuronale Netze imitieren funktionale Aspekte des Gehirns für Aufgaben wie Klassifikation und Mustererkennung. Neuere Ansätze der CI wie Reinforcement Learning ermöglichen, das Verhalten künstlicher Agenten in unbekannten Umgebungen zu steuern. Die Schwarmintelligenz modelliert Algorithmen, die auf Basis vieler einfacher Komponenten intelligente Leistungen vollführen. Zu guter Letzt lösen künstliche Immunsysteme eine Reihe von Problemen, ähnlich wie ihr biologisches Pendant. Ein kompakter und übersichtlicher mit vielen Beispielen gespickter Einstieg in die verschiedenen Verfahren der CI.