Space-filling curves (SFCs) represent an efficient and straightforward method for sparse-space indexing to transform an n-dimensional space into a one-dimensional representation. This is often applied for multidimensional point indexing which brings a better perspective for data analysis, visualization and queries. SFCs are involved in many areas such as big data analysis and visualization, image decomposition, computer graphics and geographic information systems (GISs). The indexing methods subdivide the space into logic clusters of close points and they differ in various parameters including the cluster order, the distance metrics, and the pattern shape. Beside the simple and highly preferred triangular and square uniform grids, the hexagonal uniform grids have gained high interest especially in areas such as GISs, image processing and data visualization for the uniform distance between cells and high effectiveness of circle coverage. While the linearization of hexagons is an obvious approach for memory representation, it seems there is no hexagonal SFC indexing method generally used in practice. The main limitation of hexagons lies in lacking infinite decomposition into sub-hexagons and similarity of tiles on different levels of hierarchy. Our research aims at defining a fast and robust hexagonal SFC method. The Gosper fractal is utilized to preserve the benefits of hexagonal grids and to efficiently and hierarchically linearize points in a hexagonal grid while solving the non-convex shape and recursive transformation issues of the fractal. A comparison to other SFCs and grids is conducted to verify the robustness and effectiveness of our hexagonal method.
9 June 2021 in IEEE Transactions on Evolutionary Computation
Differential evolution is a powerful nature-inspired real-parameter optimization algorithm that has been successfully used to solve a number of hard optimization problems. It has been used to tackle both continuous and discrete optimization problems. The application of a continuous method to discrete problems involves several challenges, including solution representation and search space–solution space mapping. In this work, we study random key encoding, a popular encoding scheme that is used to represent permutations in high-dimensional continuous spaces. We analyze the search space it constitutes, study its structure and properties, and introduce two novel modifications of the encoding. We investigate the proposed encoding strategies in the context of four variants of the differential evolution algorithm and demonstrate their usefulness for two widespread permutation problems: 1) the linear ordering problem and 2) the traveling salesman problem.
The Differential Evolution (DE) is a widely used bioinspired optimization algorithm developed by Storn and Price. It is popular for its simplicity and robustness. This algorithm was primarily designed for real-valued problems and continuous functions, but several modified versions optimizing both integer and discrete-valued problems have been developed. The discrete-coded DE has been mostly used for combinatorial problems in a set of enumerative variants. However, the DE has a great potential in the spatial data analysis and pattern recognition. This paper formulates the problem as a search of a combination of distinct vertices which meet the specified conditions. It proposes a novel approach called the Multidimensional Discrete Differential Evolution (MDDE) applying the principle of the discrete-coded DE in discrete point clouds (PCs). The paper examines the local searching abilities of the MDDE and its convergence to the global optimum in the PCs. The multidimensional discrete vertices cannot be simply ordered to get a convenient course of the discrete data, which is crucial for good convergence of a population. A novel mutation operator utilizing linear ordering of spatial data based on the space filling curves is introduced. The algorithm is tested on several spatial datasets and optimization problems. The experiments show that the MDDE is an efficient and fast method for discrete optimizations in the multidimensional point clouds.
The general classification is a machine learning task that tries to assign the best class to a given unknown input vector based on past observations (training data). Most of developed algorithms are very time consuming for large datasets (Support Vector Machine, Deep Neural Networks, etc.). Extreme Learning Machine (ELM) is a high quality classification algorithm that gains much popularity in recent years. This paper shows that the speed of learning of this algorithm may be improved by using GPU platform. Experimental results showed that proposed approach is much faster and provides the same accuracy as the original ELM algorithm. The proposed approach runs completely on GPU platform and thus it may be effectively incorporated within other applications.