Bogdan Filipič

DEMO: differential evolution for multiobjective optimization

Differential Evolution (DE) is a simple but powerful evolutionary optimization algorithm with many successful applications. In this paper we propose Differential Evolution for Multiobjective Optimization (DEMO) – a new approach to multiobjective optimization based on DE. DEMO combines the advantages of DE with the mechanisms of Pareto-based ranking and crowding distance sorting, used by state-of-the-art evolutionary algorithms for multiobjective optimization. DEMO is implemented in three variants that achieve competitive results on five ZDT test problems.

Noisy optimization problems - a particular challenge for differential evolution?

The popularity of search heuristics has lead to numerous new approaches in the last two decades. Since algorithm performance is problem dependent and parameter sensitive, it is difficult to consider any single approach as of greatest utility overall problems. In contrast, differential evolution (DE) is a numerical optimization approach that requires hardly any parameter tuning and is very efficient and reliable on both benchmark and real-world problems. However, the results presented in this paper demonstrate that standard methods of evolutionary optimization are able to outperform DE on noisy problems when the fitness of candidate solutions approaches the fitness variance caused by the noise.

Differential evolution versus genetic algorithms in multiobjective optimization

This paper presents a comprehensive comparison between the performance of state-of-the-art genetic algorithms NSGA-II, SPEA2 and IBEA and their differential evolution based variants DEMONS-II, DEMOSP2 and DEMOIB. Experimental results on 16 numerical multi-objective test problems show that on the majority of problems, the algorithms based on differential evolution perform significantly better than the corresponding genetic algorithms with regard to applied quality indicators. This suggests that in numerical multiobjective optimization, differential evolution explores the decision space more efficiently than genetic algorithms.

Spin label EPR-based characterization of biosystem complexity

Following the widely spread EPR spin-label applications for biosystem characterization, a novel approach is proposed for EPR-based characterization of biosystem complexity. Hereto a computational method based on a hybrid evolutionary optimization (HEO) is introduced. The enormous volume of information obtained from multiple HEO runs is reduced with a novel so-called GHOST condensation method for automatic detection of the degree of system complexity through the construction of two-dimensional solution distributions. The GHOST method shows the ability of automatic quantitative characterization of groups of solutions, e.g. the determination of average spectral parameters and group contributions. The application of the GHOST condensation algorithm is demonstrated on four synthetic examples of different complexity and applied to two physiologically relevant examples--the determination of domains in biomembranes (lateral heterogeneity) and the study of the low-resolution structure of membrane proteins.

Visualization of Pareto Front Approximations in Evolutionary Multiobjective Optimization: A Critical Review and the Prosection Method

In evolutionary multiobjective optimization, it is very important to be able to visualize approximations of the Pareto front (called approximation sets) that are found by multiobjective evolutionary algorithms. While scatter plots can be used for visualizing 2-D and 3-D approximation sets, more advanced approaches are needed to handle four or more objectives. This paper presents a comprehensive review of the existing visualization methods used in evolutionary multiobjective optimization, showing their outcomes on two novel 4-D benchmark approximation sets. In addition, a visualization method that uses prosection (projection of a section) to visualize 4-D approximation sets is proposed. The method reproduces the shape, range, and distribution of vectors in the observed approximation sets well and can handle multiple large approximation sets while being robust and computationally inexpensive. Even more importantly, for some vectors, the visualization with prosections preserves the Pareto dominance relation and relative closeness to reference points. The method is analyzed theoretically and demonstrated on several approximation sets.

Tuning EPR spectral parameters with a genetic algorithm

Abstract We present an evolutionary computation approach to parameter tuning in electron paramagnetic resonance (EPR) spectroscopy which is a nondestructive technique suitable for inspection of complex biological systems. Characterization of such a system is much more reliable when spectral features are extracted from a biophysical model of the system. This involves optimization of the model parameters so that the spectrum generated by the model matches the experimental EPR spectrum. Various optimization methods have been applied to this task in the past, but nowadays stochastic algorithms are used more and more often. As many single-point algorithms require time-consuming preparation of promising starting points to produce reasonable results, we have addressed the problem with population-based search strategy. We have implemented a genetic algorithm for EPR spectral parameter optimization and tested it on synthetic spectra obtained in cell membrane inspection. Preliminary numerical experiments show the new approach is beneficial in that it produces satisfactory results and reduces the time a spectroscopist spends for navigating the optimization process.

Data management in the MIRABEL smart grid system

Nowadays, Renewable Energy Sources (RES) are attracting more and more interest. Thus, many countries aim to increase the share of green energy and have to face with several challenges (e.g., balancing, storage, pricing). In this paper, we address the balancing challenge and present the MIRABEL project which aims to prototype an Energy Data Management System (EDMS) which takes benefit of flexibilities to efficiently balance energy demand and supply. The EDMS consists of millions of heterogeneous nodes that each incorporates advanced components (e.g., aggregation, forecasting, scheduling, negotiation). We describe each of these components and their interaction. Preliminary experimental results confirm the feasibility of our EDMS.

A comparative study of stochastic optimization methods in electric motor design

The efficiency of universal electric motors that are widely used in home appliances can be improved by optimizing the geometry of the rotor and the stator. Expert designers traditionally approach this task by iteratively evaluating candidate designs and improving them according to their experience. However, the existence of reliable numerical simulators and powerful stochastic optimization techniques make it possible to automate the design procedure. We present a comparative study of six stochastic optimization algorithms in designing optimal rotor and stator geometries of a universal electric motor where the primary objective is to minimize the motor power losses. We compare three methods from the domain of evolutionary computation, generational evolutionary algorithm, steady-state evolutionary algorithm and differential evolution, two particle-based methods, particle-swarm optimization and electromagnetism-like algorithm, and a recently proposed multilevel ant stigmergy algorithm. By comparing their performance, the most efficient method for solving the problem is identified and an explanation of its success is offered.

GP-DEMO: Differential Evolution for Multiobjective Optimization based on Gaussian Process models

This paper proposes a novel surrogate-model-based multiobjective evolutionary algorithm called Differential Evolution for Multiobjective Optimization based on Gaussian Process models (GP-DEMO). The algorithm is based on the newly defined relations for comparing solutions under uncertainty. These relations minimize the possibility of wrongly performed comparisons of solutions due to inaccurate surrogate model approximations. The GP-DEMO algorithm was tested on several benchmark problems and two computationally expensive real-world problems. To be able to assess the results we compared them with another surrogate-model-based algorithm called Generational Evolution Control (GEC) and with the Differential Evolution for Multiobjective Optimization (DEMO). The quality of the results obtained with GP-DEMO was similar to the results obtained with DEMO, but with significantly fewer exactly evaluated solutions during the optimization process. The quality of the results obtained with GEC was lower compared to the quality gained with GP-DEMO and DEMO, mainly due to wrongly performed comparisons of the inaccurately approximated solutions.

A hybrid self-adaptive evolutionary algorithm for marker optimization in the clothing industry

The task of marker optimization in clothing production is to eliminate pieces from a work order using an optimal sequence of markers and plies, where the work order is given as a matrix of colors by sizes, markers are vectors of sizes to be laid-out and cut together, and the number of plies determines how many pieces are eliminated from the work order each time. Although the optimality of a marker sequence can be determined in several ways, we consider minimum preparation cost as a key objective in clothing production. The traditional algorithms and the simple evolutionary algorithms used in marker optimization today have relied on minimizing the number of markers, which only indirectly reduces production costs. In this paper we propose a hybrid self-adaptive evolutionary algorithm (HSA-EA) for marker optimization that improves the results of the previous algorithms and successfully deals with the objective of minimum preparation cost.