Alexander Hagg

Alexander Hagg
Hochschule Bonn-Rhein-Sieg · Department of Electrical Engineering, Mechanical Engineering, and Technical Journalism (EMT)

Doctor of Philosophy
https://www.h-brs.de/en/full-domain-analysis-fluid-mechanics

About

28
Publications
3,302
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
113
Citations
Citations since 2016
25 Research Items
111 Citations
2016201720182019202020212022051015202530
2016201720182019202020212022051015202530
2016201720182019202020212022051015202530
2016201720182019202020212022051015202530
Introduction
Interested in computer aided ideation, which takes place in the first design phases. Efficient quality diversity algorithms help us to gain an intuition on the diversity of optimal solutions to an engineering design problem. I have decided to only apply my research in topics around climate adaptation, energy- and resource efficiency. Currently working on climate adapation in city planning and architecture. Open for colabs in optimization, machine learning and fluid dynamics.
Additional affiliations
March 2015 - present
Hochschule Bonn-Rhein-Sieg
Position
  • Professor
Description
  • Genetic Algorithms (B.Sc.), Neuroevolution (B.Sc.), Evolutionary Computation: Theory and Praxis (M.Sc.)
September 2013 - September 2015
Hochschule Bonn-Rhein-Sieg
Position
  • Research Assistant
Description
  • Mobile Autonomous Robots. An introductory course to research fields in robotics with a strong ROS-based programming course.
January 2011 - September 2013
Hochschule Bonn-Rhein-Sieg
Position
  • Research Assistant
Description
  • Leading exercises and tutoring in Computer Science theory and (linear/abstract) algebra.
Education
June 2017 - December 2020
Leiden University
Field of study
  • Computer Aided Ideation
January 2013 - June 2016
Hochschule Bonn-Rhein-Sieg
Field of study
  • Autonomous Systems
September 2009 - January 2013
Hochschule Bonn-Rhein-Sieg
Field of study
  • Computer Science

Publications

Publications (28)
Article
Full-text available
Current object recognition methods fail on object sets that include both diffuse, reflective and transparent materials, although they are very common in domestic scenarios. We show that a combination of cues from multiple sensor modalities, including specular reflectance and unavailable depth information, allows us to capture a larger subset of hou...
Conference Paper
Full-text available
An evolving strategy for a multi-stage placement of charging stations for electrical cars is developed. Both an incremental as well as a decremental placement decomposition are evaluated on this Maximum Covering Location Problem. We show that an incremental Genetic Algorithm benefits from problem decomposition effects of having multiple stages and...
Conference Paper
Full-text available
Evolutionary computation and genetic algorithms (GAs) in particular have been applied very successfully to many real world application problems. However, the success or failure of applying Genetic Algorithms is highly dependent on how a problem is represented. Additionally, the number of free parameters makes applying these methods a science of its...
Article
Full-text available
This paper explores the role of artificial intelligence (AI) in elite sports. We approach the topic from two perspectives. Firstly, we provide a literature based overview of AI success stories in areas other than sports. We identified multiple approaches in the area of Machine Perception, Machine Learning and Modeling, Planning and Optimization as...
Thesis
Full-text available
In this thesis, the ideas of Guilford about divergent thinking, Jung on intuition and Sartre on reflection by others are combined to create a Hegelian creative process. It is posed that the central object of preference discovery is a co-creative process in which the Other can be represented by a machine, as is often done in the computational creati...
Chapter
Here we describe quality diversity algorithms, a recent and powerful class of evolutionary algorithms that produces a diverse set of high-performing solutions. The optimization paradigm emphasizes phenotypic niching and egalitarian treatment of quality and diversity. We ground quality diversity in ecology, describe the historical development, and g...
Preprint
Full-text available
We consider multi-solution optimization and generative models for the generation of diverse artifacts and the discovery of novel solutions. In cases where the domain's factors of variation are unknown or too complex to encode manually, generative models can provide a learned latent space to approximate these factors. When used as a search space, ho...
Preprint
Full-text available
In complex, expensive optimization domains we often narrowly focus on finding high performing solutions, instead of expanding our understanding of the domain itself. But what if we could quickly understand the complex behaviors that can emerge in said domains instead? We introduce surrogate-assisted phenotypic niching, a quality diversity algorithm...
Chapter
Full-text available
More and more, optimization methods are used to find diverse solution sets. We compare solution diversity in multi-objective optimization, multimodal optimization, and quality diversity in a simple domain. We show that multiobjective optimization does not always produce much diversity, multimodal optimization produces higher fitness solutions, and...
Chapter
Full-text available
In complex, expensive optimization domains we often narrowly focus on finding high performing solutions, instead of expanding our understanding of the domain itself. But what if we could quickly understand the complex behaviors that can emerge in said domains instead? We introduce surrogate-assisted phenotypic niching, a quality diversity algorithm...
Article
Full-text available
Computers can help us to trigger our intuition about how to solve a problem. But how does a computer take into account what a user wants and update these triggers? User preferences are hard to model as they are by nature vague, depend on the user's background and are not always deterministic, changing depending on the context and process under whic...
Preprint
Full-text available
In complex, expensive optimization domains we often nar- rowly focus on finding high performing solutions, instead of expanding our understanding of the domain itself. But what if we could quickly un- derstand the complex behaviors that can emerge in said domains instead? We introduce surrogate-assisted phenotypic niching, a quality diversity algor...
Preprint
Full-text available
In optimization methods that return diverse solution sets, three interpretations of diversity can be distinguished: multi-objective optimization which searches diversity in objective space, multimodal optimization which tries spreading out the solutions in genetic space, and quality diversity which performs diversity maintenance in phenotypic space...
Technical Report
Full-text available
Aeromat Project Summary (German)
Preprint
Full-text available
Surrogate models are used to reduce the burden of expensive-to-evaluate objective functions in optimization. By creating models which map genomes to objective values, these models can estimate the performance of unknown inputs, and so be used in place of expensive objective functions. Evolutionary techniques such as genetic programming or neuroevol...
Conference Paper
Full-text available
The initial phase in real world engineering optimization and design is a process of discovery in which not all requirements can be made in advance, or are hard to formalize. Quality diversity algorithms, which produce a variety of high performing solutions, provide a unique chance to support engineers and designers in the search for what is possibl...
Conference Paper
Full-text available
Surrogate models are used to reduce the burden of expensive-to-evaluate objective functions in optimization. By creating models which map genomes to objective values, these models can estimate the performance of unknown inputs, and so be used in place of expensive objective functions. Evolutionary techniques such as genetic programming or neuroevol...
Preprint
Full-text available
The initial phase in real world engineering optimization and design is a process of discovery in which not all requirements can be made in advance, or are hard to formalize. Quality diversity algorithms, which produce a variety of high performing solutions, provide a unique chance to support engineers and designers in the search for what is possibl...
Preprint
Full-text available
An iterative computer-aided ideation procedure is introduced, building on recent quality-diversity algorithms, which search for diverse as well as high-performing solutions. Dimensionality reduction is used to define a similarity space, in which solutions are clustered into classes. These classes are represented by prototypes, which are presented t...
Conference Paper
Full-text available
Current object recognition methods fail on object sets that include both diffuse, reflective and transparent materials, although they are very common in domestic scenarios. We show that a combination of cues from multiple sensor modalities, including specular reflectance and unavailable depth information, allows us to capture a larger subset of hou...
Article
Full-text available
Maximal covering location problems have efficiently been solved using evolutionary computation. The multi-stage placement of charging stations for electric cars is an instance of this problem which is addressed in this study. It is particularly challenging, because a final solution is constructed in multiple steps, stations cannot be relocated easi...
Conference Paper
Full-text available
Evolutionary illumination is a recent technique that allows producing many diverse, optimal solutions in a map of manually defined features. To support the large amount of objective function evaluations, surrogate model assistance was recently introduced. Illumination models need to represent many more, diverse optimal regions than classical surrog...
Conference Paper
Neuroevolution methods evolve the weights of a neural network, and in some cases the topology, but little work has been done to analyze the effect of evolving the activation functions of individual nodes on network size, an important factor when training networks with a small number of samples. In this work we extend the neuroevolution algorithm NE...
Article
Full-text available
Evolutionary illumination is a recent technique that allows producing many diverse, optimal solutions in a map of manually defined features. To support the large amount of objective function evaluations, surrogate model assistance was recently introduced. Illumination models need to represent many more, diverse optimal regions than classical surrog...
Article
Full-text available
Neuroevolution methods evolve the weights of a neural network, and in some cases the topology, but little work has been done to analyze the effect of evolving the activation functions of individual nodes on network size, which is important when training networks with a small number of samples. In this work we extend the neuroevolution algorithm NEA...

Network

Cited By

Projects

Projects (5)
Project
The aim of the project is to apply AI-driven novel optimization algorithms and machine learning methods at different stages of decision-making processes in climate-adaptive urban design to model the late consequences of decisions.
Project
An Interactive Model for Real World Computational Co-creativity The aim of this project is to use divergent search methods in interactive optimization scenarios, especially in the first design phases. Quality Diversity (QD) algorithms help us to gain an intuition on the diversity of optimal solutions to an engineering design problem. A diverge-converge loop allows designers and engineers to explore optimal designs, gain insight, take decisions about what design path should be taken by the optimization algorithm and converge towards a more final design. In this project we introduce methods to increase QD efficiency, both by modeling QD features as well as sampling neural representations, automatically discover QD features, interact with the user through prototyping, modeling user preferences in both genetic as well as phenotypic spaces and finally by putting this altogether into a co-creative process that is grounded in the state of the art on computational creativity.
Project
A sustainable energy future requires that we both do more with less, and that we fully exploit the renewable energy sources we have available. In this project we explore a common thread between these two approaches, developing tools to better explore and understand aerodynamic design. On the one hand our tools can be used to improve the performance of aerodynamic vehicles, and on the other improving our ability to harvest energy from wind. We develop automated methods for the design of complete aerodynamic structures, using machine-learning techniques to guide iterative experimentation with novel designs. We focus on: 1. Optimization of entire structures, rather than iterative improvement on existing designs 2. Human-machine collaborative design exploration, to discover innovative design concepts 3. Inclusion of structural mechanics and fluid structure interaction into the optimization, design, and modeling process 4. Modeling techniques to support these goals, using data-driven approaches to approximate computationally intensive techniques and simulations In particular we face challenges when creating tools which address these issues in tandem, such as: - modeling the performance of designs produced with non-traditional parameterizations - broad exploration of possible designs in computationally demanding contexts - optimization and modeling of aerodynamic and structural properties simultaneously