Search

Real-world applications typically have multiple sparse reconstruction tasks to be optimized. In order to exploit the similar sparsity pattern between different tasks, this paper establishes an evolutionary multitasking framework to simultaneously optimize multiple sparse reconstruction tasks using a single population. In the proposed method, the evolutionary algorithm aims to search the locations of nonzero components or rows instead of searching sparse vector or matrix directly. Then the within-task and between-task genetic transfer operators are employed to reinforce the exchange of genetic material belonging to the same or different tasks. The proposed method can solve multiple measurement vector problems efficiently because the length of decision vector is independent of the number of measurement vectors. Finally, a case study on hyperspectral image unmixing is investigated in an evolutionary multitasking setting. It is natural to consider a sparse unmixing problem in a homogeneous region as a task. Experiments on signal reconstruction and hyperspectral image unmixing demonstrate the effectiveness of the proposed multitasking framework for sparse reconstruction. [ABSTRACT FROM AUTHOR]

Most evolutionary optimization algorithms assume that the evaluation of the objective and constraint functions is straightforward. In solving many real-world optimization problems, however, such objective functions may not exist. Instead, computationally expensive numerical simulations or costly physical experiments must be performed for fitness evaluations. In more extreme cases, only historical data are available for performing optimization and no new data can be generated during optimization. Solving evolutionary optimization problems driven by data collected in simulations, physical experiments, production processes, or daily life are termed data-driven evolutionary optimization. In this paper, we provide a taxonomy of different data driven evolutionary optimization problems, discuss main challenges in data-driven evolutionary optimization with respect to the nature and amount of data, and the availability of new data during optimization. Real-world application examples are given to illustrate different model management strategies for different categories of data-driven optimization problems. [ABSTRACT FROM AUTHOR]

This paper investigates the issue of evolutionary design of open-ended plants for hybrid dynamical systems, i.e., both their topologies and parameters. Hybrid bond graphs (HBGs) are used to represent dynamical systems involving both continuous and discrete system dynamics. Genetic programming, with some special mechanisms incorporated, is used as a search tool to explore the open-ended design space of hybrid bond graphs. Combination of these two tools, i.e., HBGs and genetic programming, leads to an approach called HBGGP that can automatically generate viable design candidates of hybrid dynamical systems that fulfill predefined design specifications. A comprehensive investigation of a case study of DC-DC converter design demonstrates the feasibility and effectiveness of the HBGGP approach. Important characteristics of the approach are also discussed, with some future research directions pointed out. [ABSTRACT FROM PUBLISHER]

Many random events usually are associated with executions of operational plans at various companies and organizations. For example, some tasks might be delayed and/or executed earlier. Some operational constraints can be introduced due to new regulations or business rules. In some cases, there might be a shift in the relative importance of objectives associated with these plans. All these potential modifications create a huge pressure on planning staff for generating plans that can adapt quickly to changes in environment during execution. In this paper, we address adaptation in dynamic environments. Many researchers in evolutionary community addressed the problem of optimization in dynamic environments. Through an overview on applying evolutionary algorithms for solving dynamic optimization problems, we classify the paper into two main categories: 1) finding/tracking optima, and 2) adaptation and we discuss their relevance for solving planning problems. Based on this discussion, we propose a computational approach to adaptation within the context of planning. This approach models the dynamic planning problem as a multiobjective optimization problem and an evolutionary mechanism is incorporated; this adapts the current solution to new situations when a change occurs. As the multiobjective model is used, the proposed approach produces a set of non-dominated solutions after each planning cycle. This set of solutions can be perceived as an information-rich data set which can be used to support the adaptation process against the effect of changes. The main question is how to exploit this set efficiently. In this paper, we propose a method based on the concept of centroids over a number of changing-time steps; at each step we obtain a set of non-dominated solutions. We carried out a case study on this proposed approach. Mission planning was used for our experiments and experimental analysis. We selected mission planning as our test environment because battlefields are always highly dynamic and uncertain and can be conveniently used to demonstrate different types of changes, especially time-varying constraints. The obtained results support the significance of our centroid-based approach. [ABSTRACT FROM PUBLISHER]

Sesame is a software framework that aims at developing a modeling and simulation environment for the efficient design space exploration of heterogeneous embedded systems. Since Sesame recognizes separate application and architecture models within a single system simulation, it needs an explicit mapping step to relate these models for cosimulation. The design tradeoffs during the mapping stage, namely, the processing time, power consumption, and architecture cost, are captured by a multiobjective nonlinear mixed integer program. This paper aims at investigating the performance of multiobjective evolutionary algorithms (MOEAs) on solving large instances of the mapping problem. With two comparative case studies, it is shown that MOEAs provide the designer with a highly accurate set of solutions in a reasonable amount of time. Additionally, analyses for different crossover types, mutation usage, and repair strategies for the purpose of constraints handling are carried out. Finally, a number of multiobjective optimization results are simulated for verification. [ABSTRACT FROM AUTHOR]

Population-based memetic algorithms have been successfully applied to solve many difficult combinatorial problems. Often, a population of fixed size is used in such algorithms to record some best solutions sampled during the search. However, given the particular features of the problem instance under consideration, a population of variable size would be more suitable to ensure the best search performance possible. In this work, we propose a variable population memetic search (VPMS), where a strategic population sizing mechanism is used to dynamically adjust the population size during the search process. Our VPMS approach starts its search from a small population of only two solutions to focus on exploitation and then adapts the population size according to the search status to continuously influence the balancing between exploitation and exploration. We illustrate an application of the VPMS approach to solve the challenging critical node problem (CNP). We show that the VPMS algorithm integrating a variable population, an effective local optimization procedure, and a backbone-based crossover operator performs very well compared to state-of-the-art CNP algorithms. The algorithm is able to discover new upper bounds for 12 instances out of the 42 popular benchmark instances while matching 23 previous best-known upper bounds. [ABSTRACT FROM AUTHOR]