@article{21004, abstract = {{Automated machine learning (AutoML) supports the algorithmic construction and data-specific customization of machine learning pipelines, including the selection, combination, and parametrization of machine learning algorithms as main constituents. Generally speaking, AutoML approaches comprise two major components: a search space model and an optimizer for traversing the space. Recent approaches have shown impressive results in the realm of supervised learning, most notably (single-label) classification (SLC). Moreover, first attempts at extending these approaches towards multi-label classification (MLC) have been made. While the space of candidate pipelines is already huge in SLC, the complexity of the search space is raised to an even higher power in MLC. One may wonder, therefore, whether and to what extent optimizers established for SLC can scale to this increased complexity, and how they compare to each other. This paper makes the following contributions: First, we survey existing approaches to AutoML for MLC. Second, we augment these approaches with optimizers not previously tried for MLC. Third, we propose a benchmarking framework that supports a fair and systematic comparison. Fourth, we conduct an extensive experimental study, evaluating the methods on a suite of MLC problems. We find a grammar-based best-first search to compare favorably to other optimizers.}}, author = {{Wever, Marcel Dominik and Tornede, Alexander and Mohr, Felix and Hüllermeier, Eyke}}, issn = {{0162-8828}}, journal = {{IEEE Transactions on Pattern Analysis and Machine Intelligence}}, keywords = {{Automated Machine Learning, Multi Label Classification, Hierarchical Planning, Bayesian Optimization}}, pages = {{1--1}}, title = {{{AutoML for Multi-Label Classification: Overview and Empirical Evaluation}}}, doi = {{10.1109/tpami.2021.3051276}}, year = {{2021}}, } @article{46318, abstract = {{Multi-objective (MO) optimization, i.e., the simultaneous optimization of multiple conflicting objectives, is gaining more and more attention in various research areas, such as evolutionary computation, machine learning (e.g., (hyper-)parameter optimization), or logistics (e.g., vehicle routing). Many works in this domain mention the structural problem property of multimodality as a challenge from two classical perspectives: (1) finding all globally optimal solution sets, and (2) avoiding to get trapped in local optima. Interestingly, these streams seem to transfer many traditional concepts of single-objective (SO) optimization into claims, assumptions, or even terminology regarding the MO domain, but mostly neglect the understanding of the structural properties as well as the algorithmic search behavior on a problem’s landscape. However, some recent works counteract this trend, by investigating the fundamentals and characteristics of MO problems using new visualization techniques and gaining surprising insights. Using these visual insights, this work proposes a step towards a unified terminology to capture multimodality and locality in a broader way than it is usually done. This enables us to investigate current research activities in multimodal continuous MO optimization and to highlight new implications and promising research directions for the design of benchmark suites, the discovery of MO landscape features, the development of new MO (or even SO) optimization algorithms, and performance indicators. For all these topics, we provide a review of ideas and methods but also an outlook on future challenges, research potential and perspectives that result from recent developments.}}, author = {{Grimme, Christian and Kerschke, Pascal and Aspar, Pelin and Trautmann, Heike and Preuss, Mike and Deutz, André H. and Wang, Hao and Emmerich, Michael}}, issn = {{0305-0548}}, journal = {{Computers & Operations Research}}, keywords = {{Multimodal optimization, Multi-objective continuous optimization, Landscape analysis, Visualization, Benchmarking, Theory, Algorithms}}, pages = {{105489}}, title = {{{Peeking beyond peaks: Challenges and research potentials of continuous multimodal multi-objective optimization}}}, doi = {{https://doi.org/10.1016/j.cor.2021.105489}}, volume = {{136}}, year = {{2021}}, } @inproceedings{48853, abstract = {{In practise, it is often desirable to provide the decision-maker with a rich set of diverse solutions of decent quality instead of just a single solution. In this paper we study evolutionary diversity optimization for the knapsack problem (KP). Our goal is to evolve a population of solutions that all have a profit of at least (1 - {$ϵ$}) {$\cdot$} OPT, where OPT is the value of an optimal solution. Furthermore, they should differ in structure with respect to an entropy-based diversity measure. To this end we propose a simple ({$\mu$} + 1)-EA with initial approximate solutions calculated by a well-known FPTAS for the KP. We investigate the effect of different standard mutation operators and introduce biased mutation and crossover which puts strong probability on flipping bits of low and/or high frequency within the population. An experimental study on different instances and settings shows that the proposed mutation operators in most cases perform slightly inferior in the long term, but show strong benefits if the number of function evaluations is severely limited.}}, author = {{Bossek, Jakob and Neumann, Aneta and Neumann, Frank}}, booktitle = {{Proceedings of the Genetic and Evolutionary Computation Conference}}, isbn = {{978-1-4503-8350-9}}, keywords = {{evolutionary algorithms, evolutionary diversity optimization, knapsack problem, tailored operators}}, pages = {{556–564}}, publisher = {{Association for Computing Machinery}}, title = {{{Breeding Diverse Packings for the Knapsack Problem by Means of Diversity-Tailored Evolutionary Algorithms}}}, doi = {{10.1145/3449639.3459364}}, year = {{2021}}, } @inproceedings{48860, abstract = {{In the area of evolutionary computation the calculation of diverse sets of high-quality solutions to a given optimization problem has gained momentum in recent years under the term evolutionary diversity optimization. Theoretical insights into the working principles of baseline evolutionary algorithms for diversity optimization are still rare. In this paper we study the well-known Minimum Spanning Tree problem (MST) in the context of diversity optimization where population diversity is measured by the sum of pairwise edge overlaps. Theoretical results provide insights into the fitness landscape of the MST diversity optimization problem pointing out that even for a population of {$\mu$} = 2 fitness plateaus (of constant length) can be reached, but nevertheless diverse sets can be calculated in polynomial time. We supplement our theoretical results with a series of experiments for the unconstrained and constraint case where all solutions need to fulfill a minimal quality threshold. Our results show that a simple ({$\mu$} + 1)-EA can effectively compute a diversified population of spanning trees of high quality.}}, author = {{Bossek, Jakob and Neumann, Frank}}, booktitle = {{Proceedings of the Genetic and Evolutionary Computation Conference}}, isbn = {{978-1-4503-8350-9}}, keywords = {{evolutionary algorithms, evolutionary diversity optimization, minimum spanning tree, runtime analysis}}, pages = {{198–206}}, publisher = {{Association for Computing Machinery}}, title = {{{Evolutionary Diversity Optimization and the Minimum Spanning Tree Problem}}}, doi = {{10.1145/3449639.3459363}}, year = {{2021}}, } @article{48854, abstract = {{We contribute to the theoretical understanding of randomized search heuristics for dynamic problems. We consider the classical vertex coloring problem on graphs and investigate the dynamic setting where edges are added to the current graph. We then analyze the expected time for randomized search heuristics to recompute high quality solutions. The (1+1) Evolutionary Algorithm and RLS operate in a setting where the number of colors is bounded and we are minimizing the number of conflicts. Iterated local search algorithms use an unbounded color palette and aim to use the smallest colors and, consequently, the smallest number of colors. We identify classes of bipartite graphs where reoptimization is as hard as or even harder than optimization from scratch, i.e., starting with a random initialization. Even adding a single edge can lead to hard symmetry problems. However, graph classes that are hard for one algorithm turn out to be easy for others. In most cases our bounds show that reoptimization is faster than optimizing from scratch. We further show that tailoring mutation operators to parts of the graph where changes have occurred can significantly reduce the expected reoptimization time. In most settings the expected reoptimization time for such tailored algorithms is linear in the number of added edges. However, tailored algorithms cannot prevent exponential times in settings where the original algorithm is inefficient.}}, author = {{Bossek, Jakob and Neumann, Frank and Peng, Pan and Sudholt, Dirk}}, issn = {{0178-4617}}, journal = {{Algorithmica}}, keywords = {{Dynamic optimization, Evolutionary algorithms, Running time analysis}}, number = {{10}}, pages = {{3148–3179}}, title = {{{Time Complexity Analysis of Randomized Search Heuristics for the Dynamic Graph Coloring Problem}}}, doi = {{10.1007/s00453-021-00838-3}}, volume = {{83}}, year = {{2021}}, } @article{21436, abstract = {{Ultrasonic wire bonding is a solid-state joining process, used in the electronics industry to form electrical connections, e.g. to connect electrical terminals within semiconductor modules. Many process parameters affect the bond strength, such like the bond normal force, ultrasonic power, wire material and bonding frequency. Today, process design, development, and optimization is most likely based on the knowledge of process engineers and is mainly performed by experimental testing. In this contribution, a newly developed simulation tool is presented, to reduce time and costs and efficiently determine optimized process parameter. Based on a co-simulation of MATLAB and ANSYS, the different physical phenomena of the wire bonding process are considered using finite element simulation for the complex plastic deformation of the wire and reduced order models for the transient dynamics of the transducer, wire, substrate and bond formation. The model parameters such as the coefficients of friction between bond tool and wire and between wire and substrate were determined for aluminium and copper wire in experiments with a test rig specially developed for the requirements of heavy wire bonding. To reduce simulation time, for the finite element simulation a restart analysis and high performance computing is utilized. Detailed analysis of the bond formation showed, that the normal pressure distribution in the contact between wire and substrate has high impact on bond formation and distribution of welded areas in the contact area.}}, author = {{Schemmel, Reinhard and Krieger, Viktor and Hemsel, Tobias and Sextro, Walter}}, issn = {{0026-2714}}, journal = {{Microelectronics Reliability}}, keywords = {{Ultrasonic heavy wire bonding, Co-simulation, ANSYS, MATLAB, Process optimization, Friction coefficient, Copper-copper, Aluminium-copper}}, pages = {{114077}}, title = {{{Co-simulation of MATLAB and ANSYS for ultrasonic wire bonding process optimization}}}, doi = {{https://doi.org/10.1016/j.microrel.2021.114077}}, volume = {{119}}, year = {{2021}}, } @inproceedings{17370, abstract = {{ We consider a natural extension to the metric uncapacitated Facility Location Problem (FLP) in which requests ask for different commodities out of a finite set \( S \) of commodities. Ravi and Sinha (SODA 2004) introduced the model as the \emph{Multi-Commodity Facility Location Problem} (MFLP) and considered it an offline optimization problem. The model itself is similar to the FLP: i.e., requests are located at points of a finite metric space and the task of an algorithm is to construct facilities and assign requests to facilities while minimizing the construction cost and the sum over all assignment distances. In addition, requests and facilities are heterogeneous; they request or offer multiple commodities out of $S$. A request has to be connected to a set of facilities jointly offering the commodities demanded by it. In comparison to the FLP, an algorithm has to decide not only if and where to place facilities, but also which commodities to offer at each. To the best of our knowledge we are the first to study the problem in its online variant in which requests, their positions and their commodities are not known beforehand but revealed over time. We present results regarding the competitive ratio. On the one hand, we show that heterogeneity influences the competitive ratio by developing a lower bound on the competitive ratio for any randomized online algorithm of \( \Omega ( \sqrt{|S|} + \frac{\log n}{\log \log n} ) \) that already holds for simple line metrics. Here, \( n \) is the number of requests. On the other side, we establish a deterministic \( \mathcal{O}(\sqrt{|S|} \cdot \log n) \)-competitive algorithm and a randomized \( \mathcal{O}(\sqrt{|S|} \cdot \frac{\log n}{\log \log n} ) \)-competitive algorithm. Further, we show that when considering a more special class of cost functions for the construction cost of a facility, the competitive ratio decreases given by our deterministic algorithm depending on the function.}}, author = {{Castenow, Jannik and Feldkord, Björn and Knollmann, Till and Malatyali, Manuel and Meyer auf der Heide, Friedhelm}}, booktitle = {{Proceedings of the 32nd ACM Symposium on Parallelism in Algorithms and Architectures}}, isbn = {{9781450369350}}, keywords = {{Online Multi-Commodity Facility Location, Competitive Ratio, Online Optimization, Facility Location Problem}}, title = {{{The Online Multi-Commodity Facility Location Problem}}}, doi = {{10.1145/3350755.3400281}}, year = {{2020}}, } @inproceedings{48847, abstract = {{Dynamic optimization problems have gained significant attention in evolutionary computation as evolutionary algorithms (EAs) can easily adapt to changing environments. We show that EAs can solve the graph coloring problem for bipartite graphs more efficiently by using dynamic optimization. In our approach the graph instance is given incrementally such that the EA can reoptimize its coloring when a new edge introduces a conflict. We show that, when edges are inserted in a way that preserves graph connectivity, Randomized Local Search (RLS) efficiently finds a proper 2-coloring for all bipartite graphs. This includes graphs for which RLS and other EAs need exponential expected time in a static optimization scenario. We investigate different ways of building up the graph by popular graph traversals such as breadth-first-search and depth-first-search and analyse the resulting runtime behavior. We further show that offspring populations (e. g. a (1 + {$\lambda$}) RLS) lead to an exponential speedup in {$\lambda$}. Finally, an island model using 3 islands succeeds in an optimal time of {$\Theta$}(m) on every m-edge bipartite graph, outperforming offspring populations. This is the first example where an island model guarantees a speedup that is not bounded in the number of islands.}}, author = {{Bossek, Jakob and Neumann, Frank and Peng, Pan and Sudholt, Dirk}}, booktitle = {{Proceedings of the Genetic and Evolutionary Computation Conference}}, isbn = {{978-1-4503-7128-5}}, keywords = {{dynamic optimization, evolutionary algorithms, running time analysis, theory}}, pages = {{1277–1285}}, publisher = {{Association for Computing Machinery}}, title = {{{More Effective Randomized Search Heuristics for Graph Coloring through Dynamic Optimization}}}, doi = {{10.1145/3377930.3390174}}, year = {{2020}}, } @inproceedings{48849, abstract = {{One-shot optimization tasks require to determine the set of solution candidates prior to their evaluation, i.e., without possibility for adaptive sampling. We consider two variants, classic one-shot optimization (where our aim is to find at least one solution of high quality) and one-shot regression (where the goal is to fit a model that resembles the true problem as well as possible). For both tasks it seems intuitive that well-distributed samples should perform better than uniform or grid-based samples, since they show a better coverage of the decision space. In practice, quasi-random designs such as Latin Hypercube Samples and low-discrepancy point sets are indeed very commonly used designs for one-shot optimization tasks. We study in this work how well low star discrepancy correlates with performance in one-shot optimization. Our results confirm an advantage of low-discrepancy designs, but also indicate the correlation between discrepancy values and overall performance is rather weak. We then demonstrate that commonly used designs may be far from optimal. More precisely, we evolve 24 very specific designs that each achieve good performance on one of our benchmark problems. Interestingly, we find that these specifically designed samples yield surprisingly good performance across the whole benchmark set. Our results therefore give strong indication that significant performance gains over state-of-the-art one-shot sampling techniques are possible, and that evolutionary algorithms can be an efficient means to evolve these.}}, author = {{Bossek, Jakob and Doerr, Carola and Kerschke, Pascal and Neumann, Aneta and Neumann, Frank}}, booktitle = {{Parallel Problem Solving from Nature (PPSN XVI)}}, isbn = {{978-3-030-58111-4}}, keywords = {{Continuous optimization, Fully parallel search, One-shot optimization, Regression, Surrogate-assisted optimization}}, pages = {{111–124}}, publisher = {{Springer-Verlag}}, title = {{{Evolving Sampling Strategies for One-Shot Optimization Tasks}}}, doi = {{10.1007/978-3-030-58112-1_8}}, year = {{2020}}, } @inproceedings{48851, abstract = {{Several important optimization problems in the area of vehicle routing can be seen as variants of the classical Traveling Salesperson Problem (TSP). In the area of evolutionary computation, the Traveling Thief Problem (TTP) has gained increasing interest over the last 5 years. In this paper, we investigate the effect of weights on such problems, in the sense that the cost of traveling increases with respect to the weights of nodes already visited during a tour. This provides abstractions of important TSP variants such as the Traveling Thief Problem and time dependent TSP variants, and allows to study precisely the increase in difficulty caused by weight dependence. We provide a 3.59-approximation for this weight dependent version of TSP with metric distances and bounded positive weights. Furthermore, we conduct experimental investigations for simple randomized local search with classical mutation operators and two variants of the state-of-the-art evolutionary algorithm EAX adapted to the weighted TSP. Our results show the impact of the node weights on the position of the nodes in the resulting tour.}}, author = {{Bossek, Jakob and Casel, Katrin and Kerschke, Pascal and Neumann, Frank}}, booktitle = {{Proceedings of the Genetic and Evolutionary Computation Conference}}, isbn = {{978-1-4503-7128-5}}, keywords = {{dynamic optimization, evolutionary algorithms, running time analysis, theory}}, pages = {{1286–1294}}, publisher = {{Association for Computing Machinery}}, title = {{{The Node Weight Dependent Traveling Salesperson Problem: Approximation Algorithms and Randomized Search Heuristics}}}, doi = {{10.1145/3377930.3390243}}, year = {{2020}}, } @inproceedings{48845, abstract = {{In practice, e.g. in delivery and service scenarios, Vehicle-Routing-Problems (VRPs) often imply repeated decision making on dynamic customer requests. As in classical VRPs, tours have to be planned short while the number of serviced customers has to be maximized at the same time resulting in a multi-objective problem. Beyond that, however, dynamic requests lead to the need for re-planning of not yet realized tour parts, while already realized tour parts are irreversible. In this paper we study this type of bi-objective dynamic VRP including sequential decision making and concurrent realization of decisions. We adopt a recently proposed Dynamic Evolutionary Multi-Objective Algorithm (DEMOA) for a related VRP problem and extend it to the more realistic (here considered) scenario of multiple vehicles. We empirically show that our DEMOA is competitive with a multi-vehicle offline and clairvoyant variant of the proposed DEMOA as well as with the dynamic single-vehicle approach proposed earlier.}}, author = {{Bossek, Jakob and Grimme, Christian and Trautmann, Heike}}, booktitle = {{Proceedings of the Genetic and Evolutionary Computation Conference}}, isbn = {{978-1-4503-7128-5}}, keywords = {{decision making, dynamic optimization, evolutionary algorithms, multi-objective optimization, vehicle routing}}, pages = {{166–174}}, publisher = {{Association for Computing Machinery}}, title = {{{Dynamic Bi-Objective Routing of Multiple Vehicles}}}, doi = {{10.1145/3377930.3390146}}, year = {{2020}}, } @inproceedings{48850, abstract = {{Sequential model-based optimization (SMBO) approaches are algorithms for solving problems that require computationally or otherwise expensive function evaluations. The key design principle of SMBO is a substitution of the true objective function by a surrogate, which is used to propose the point(s) to be evaluated next. SMBO algorithms are intrinsically modular, leaving the user with many important design choices. Significant research efforts go into understanding which settings perform best for which type of problems. Most works, however, focus on the choice of the model, the acquisition function, and the strategy used to optimize the latter. The choice of the initial sampling strategy, however, receives much less attention. Not surprisingly, quite diverging recommendations can be found in the literature. We analyze in this work how the size and the distribution of the initial sample influences the overall quality of the efficient global optimization (EGO) algorithm, a well-known SMBO approach. While, overall, small initial budgets using Halton sampling seem preferable, we also observe that the performance landscape is rather unstructured. We furthermore identify several situations in which EGO performs unfavorably against random sampling. Both observations indicate that an adaptive SMBO design could be beneficial, making SMBO an interesting test-bed for automated algorithm design.}}, author = {{Bossek, Jakob and Doerr, Carola and Kerschke, Pascal}}, booktitle = {{Proceedings of the Genetic and Evolutionary Computation Conference}}, isbn = {{978-1-4503-7128-5}}, keywords = {{continuous black-box optimization, design of experiments, initial design, sequential model-based optimization}}, pages = {{778–786}}, publisher = {{Association for Computing Machinery}}, title = {{{Initial Design Strategies and Their Effects on Sequential Model-Based Optimization: An Exploratory Case Study Based on BBOB}}}, doi = {{10.1145/3377930.3390155}}, year = {{2020}}, } @article{48848, abstract = {{We build upon a recently proposed multi-objective view onto performance measurement of single-objective stochastic solvers. The trade-off between the fraction of failed runs and the mean runtime of successful runs \textendash both to be minimized \textendash is directly analyzed based on a study on algorithm selection of inexact state-of-the-art solvers for the famous Traveling Salesperson Problem (TSP). Moreover, we adopt the hypervolume indicator (HV) commonly used in multi-objective optimization for simultaneously assessing both conflicting objectives and investigate relations to commonly used performance indicators, both theoretically and empirically. Next to Penalized Average Runtime (PAR) and Penalized Quantile Runtime (PQR), the HV measure is used as a core concept within the construction of per-instance algorithm selection models offering interesting insights into complementary behavior of inexact TSP solvers. \textbullet The multi-objective perspective is naturally generalizable to multiple objectives. \textbullet Proof of relationship between HV and the PAR in the considered bi-objective space. \textbullet New insights into complementary behavior of stochastic optimization algorithms.}}, author = {{Bossek, Jakob and Kerschke, Pascal and Trautmann, Heike}}, issn = {{1568-4946}}, journal = {{Applied Soft Computing}}, keywords = {{Algorithm selection, Combinatorial optimization, Multi-objective optimization, Performance measurement, Traveling Salesperson Problem}}, number = {{C}}, title = {{{A Multi-Objective Perspective on Performance Assessment and Automated Selection of Single-Objective Optimization Algorithms}}}, doi = {{10.1016/j.asoc.2019.105901}}, volume = {{88}}, year = {{2020}}, } @article{46334, abstract = {{We build upon a recently proposed multi-objective view onto performance measurement of single-objective stochastic solvers. The trade-off between the fraction of failed runs and the mean runtime of successful runs – both to be minimized – is directly analyzed based on a study on algorithm selection of inexact state-of-the-art solvers for the famous Traveling Salesperson Problem (TSP). Moreover, we adopt the hypervolume indicator (HV) commonly used in multi-objective optimization for simultaneously assessing both conflicting objectives and investigate relations to commonly used performance indicators, both theoretically and empirically. Next to Penalized Average Runtime (PAR) and Penalized Quantile Runtime (PQR), the HV measure is used as a core concept within the construction of per-instance algorithm selection models offering interesting insights into complementary behavior of inexact TSP solvers.}}, author = {{Bossek, Jakob and Kerschke, Pascal and Trautmann, Heike}}, issn = {{1568-4946}}, journal = {{Applied Soft Computing}}, keywords = {{Algorithm selection, Multi-objective optimization, Performance measurement, Combinatorial optimization, Traveling Salesperson Problem}}, pages = {{105901}}, title = {{{A multi-objective perspective on performance assessment and automated selection of single-objective optimization algorithms}}}, doi = {{https://doi.org/10.1016/j.asoc.2019.105901}}, volume = {{88}}, year = {{2020}}, } @article{33866, abstract = {{Helhmoltz–Kirchhoff equations of motions of vortices of an incompressible fluid in the plane define a dynamics with singularities and this leads to a Zermelo navigation problem describing the ship travel in such a field where the control is the heading angle. Considering one vortex, we define a time minimization problem which can be analyzed with the technics of geometric optimal control combined with numerical simulations, the geometric frame being the extension of Randers metrics in the punctured plane, with rotational symmetry. Candidates as minimizers are parameterized thanks to the Pontryagin Maximum Principle as extremal solutions of a Hamiltonian vector field. We analyze the time minimal solution to transfer the ship between two points where during the transfer the ship can be either in a strong current region in the vicinity of the vortex or in a weak current region. The analysis is based on a micro-local classification of the extremals using mainly the integrability properties of the dynamics due to the rotational symmetry. The discussion is complex and related to the existence of an isolated extremal (Reeb) circle due to the vortex singularity. The explicit computation of cut points where the extremal curves cease to be optimal is given and the spheres are described in the case where at the initial point the current is weak.}}, author = {{Bonnard, Bernard and Cots, Olivier and Wembe Moafo, Boris Edgar}}, issn = {{1292-8119}}, journal = {{ESAIM: Control, Optimisation and Calculus of Variations}}, keywords = {{Computational Mathematics, Control and Optimization, Control and Systems Engineering}}, publisher = {{EDP Sciences}}, title = {{{A Zermelo navigation problem with a vortex singularity}}}, doi = {{10.1051/cocv/2020058}}, volume = {{27}}, year = {{2020}}, } @article{4562, abstract = {{Employing main and sector-specific investment-grade CDS indices from the North American and European CDS market and performing mean-variance out-of-sample analyses for conservative and aggressive investors over the period from 2006 to 2014, this paper analyzes portfolio benefits of adding corporate CDS indices to a traditional financial portfolio consisting of stock and sovereign bond indices. As a baseline result, we initially find an increase in portfolio (downside) risk-diversification when adding CDS indices, which is observed irrespective of both CDS markets, investor-types and different sub-periods, including the global financial crisis and European sovereign debt crisis. In addition, the analysis reveals higher portfolio excess returns and performance in CDS index portfolios, however, these effects clearly differ between markets, investor-types and sub-periods. Overall, portfolio benefits of adding CDS indices mainly result from the fact that institutional investors replace sovereign bond indices rather than stock indices by CDS indices due to better risk-return characteristics. Our baseline findings remain robust under a variety of robustness checks. Results from sensitivity analyses provide further important implications for institutional investors with a strategic focus on a long-term conservative portfolio management.}}, author = {{Hippert, Benjamin and Uhde, André and Wengerek, Sascha Tobias}}, journal = {{Review of Derivatives Research }}, keywords = {{Corporate credit default swap indices, Mean-variance asset allocation, Out-of-sample portfolio optimization, Portfolio risk-diversification, Portfolio performance evaluation}}, number = {{2}}, pages = {{203--259}}, title = {{{Portfolio Benefits of Adding Corporate Credit Default Swap Indices: Evidence from North America and Europe}}}, doi = {{https://doi.org/10.1007/s11147-018-9148-8}}, volume = {{22}}, year = {{2019}}, } @inproceedings{48841, abstract = {{We tackle a bi-objective dynamic orienteering problem where customer requests arise as time passes by. The goal is to minimize the tour length traveled by a single delivery vehicle while simultaneously keeping the number of dismissed dynamic customers to a minimum. We propose a dynamic Evolutionary Multi-Objective Algorithm which is grounded on insights gained from a previous series of work on an a-posteriori version of the problem, where all request times are known in advance. In our experiments, we simulate different decision maker strategies and evaluate the development of the Pareto-front approximations on exemplary problem instances. It turns out, that despite severely reduced computational budget and no oracle-knowledge of request times the dynamic EMOA is capable of producing approximations which partially dominate the results of the a-posteriori EMOA and dynamic integer linear programming strategies.}}, author = {{Bossek, Jakob and Grimme, Christian and Meisel, Stephan and Rudolph, Günter and Trautmann, Heike}}, booktitle = {{Evolutionary Multi-Criterion Optimization (EMO)}}, editor = {{Deb, Kalyanmoy and Goodman, Erik and Coello Coello, Carlos A. and Klamroth, Kathrin and Miettinen, Kaisa and Mostaghim, Sanaz and Reed, Patrick}}, isbn = {{978-3-030-12598-1}}, keywords = {{Combinatorial optimization, Dynamic optimization, Metaheuristics, Multi-objective optimization, Vehicle routing}}, pages = {{516–528}}, publisher = {{Springer International Publishing}}, title = {{{Bi-Objective Orienteering: Towards a Dynamic Multi-objective Evolutionary Algorithm}}}, doi = {{10.1007/978-3-030-12598-1_41}}, year = {{2019}}, } @inproceedings{48842, abstract = {{Evolutionary algorithms have successfully been applied to evolve problem instances that exhibit a significant difference in performance for a given algorithm or a pair of algorithms inter alia for the Traveling Salesperson Problem (TSP). Creating a large variety of instances is crucial for successful applications in the blooming field of algorithm selection. In this paper, we introduce new and creative mutation operators for evolving instances of the TSP. We show that adopting those operators in an evolutionary algorithm allows for the generation of benchmark sets with highly desirable properties: (1) novelty by clear visual distinction to established benchmark sets in the field, (2) visual and quantitative diversity in the space of TSP problem characteristics, and (3) significant performance differences with respect to the restart versions of heuristic state-of-the-art TSP solvers EAX and LKH. The important aspect of diversity is addressed and achieved solely by the proposed mutation operators and not enforced by explicit diversity preservation.}}, author = {{Bossek, Jakob and Kerschke, Pascal and Neumann, Aneta and Wagner, Markus and Neumann, Frank and Trautmann, Heike}}, booktitle = {{Proceedings of the 15th ACM/SIGEVO Conference on Foundations of Genetic Algorithms}}, isbn = {{978-1-4503-6254-2}}, keywords = {{benchmarking, instance features, optimization, problem generation, traveling salesperson problem}}, pages = {{58–71}}, publisher = {{Association for Computing Machinery}}, title = {{{Evolving Diverse TSP Instances by Means of Novel and Creative Mutation Operators}}}, doi = {{10.1145/3299904.3340307}}, year = {{2019}}, } @inproceedings{48843, abstract = {{We contribute to the theoretical understanding of randomized search heuristics for dynamic problems. We consider the classical graph coloring problem and investigate the dynamic setting where edges are added to the current graph. We then analyze the expected time for randomized search heuristics to recompute high quality solutions. This includes the (1+1) EA and RLS in a setting where the number of colors is bounded and we are minimizing the number of conflicts as well as iterated local search algorithms that use an unbounded color palette and aim to use the smallest colors and - as a consequence - the smallest number of colors. We identify classes of bipartite graphs where reoptimization is as hard as or even harder than optimization from scratch, i. e. starting with a random initialization. Even adding a single edge can lead to hard symmetry problems. However, graph classes that are hard for one algorithm turn out to be easy for others. In most cases our bounds show that reoptimization is faster than optimizing from scratch. Furthermore, we show how to speed up computations by using problem specific operators concentrating on parts of the graph where changes have occurred.}}, author = {{Bossek, Jakob and Neumann, Frank and Peng, Pan and Sudholt, Dirk}}, booktitle = {{Proceedings of the Genetic and Evolutionary Computation Conference}}, isbn = {{978-1-4503-6111-8}}, keywords = {{dynamic optimization, evolutionary algorithms, running time analysis, theory}}, pages = {{1443–1451}}, publisher = {{Association for Computing Machinery}}, title = {{{Runtime Analysis of Randomized Search Heuristics for Dynamic Graph Coloring}}}, doi = {{10.1145/3321707.3321792}}, year = {{2019}}, } @inproceedings{48840, abstract = {{Research has shown that for many single-objective graph problems where optimum solutions are composed of low weight sub-graphs, such as the minimum spanning tree problem (MST), mutation operators favoring low weight edges show superior performance. Intuitively, similar observations should hold for multi-criteria variants of such problems. In this work, we focus on the multi-criteria MST problem. A thorough experimental study is conducted where we estimate the probability of edges being part of non-dominated spanning trees as a function of the edges’ non-domination level or domination count, respectively. Building on gained insights, we propose several biased one-edge-exchange mutation operators that differ in the used edge-selection probability distribution (biased towards edges of low rank). Our empirical analysis shows that among different graph types (dense and sparse) and edge weight types (both uniformly random and combinations of Euclidean and uniformly random) biased edge-selection strategies perform superior in contrast to the baseline uniform edge-selection. Our findings are in particular strong for dense graphs.}}, author = {{Bossek, Jakob and Grimme, Christian and Neumann, Frank}}, booktitle = {{Proceedings of the Genetic and Evolutionary Computation Conference}}, isbn = {{978-1-4503-6111-8}}, keywords = {{biased mutation, combinatorial optimization, minimum spanning tree, multi-objective optimization}}, pages = {{516–523}}, publisher = {{Association for Computing Machinery}}, title = {{{On the Benefits of Biased Edge-Exchange Mutation for the Multi-Criteria Spanning Tree Problem}}}, doi = {{10.1145/3321707.3321818}}, year = {{2019}}, }