In this paper, we incorporate a local search procedure into a micro differential evolution algorithm MED with the aim of solving the HappyCat function. Our purpose is to find out if our proposal is more competitive than a Ray-ES algorithm. We test our micro Differential Evolution algorithm (µDE) on HappyCat and HGBat functions. The results that we obtained with micro-DE are better compared with the results the original RayES reference algorithm. This analysis supports our conjecture that a reduced population DE hybridized with a local search (Ray search) is a key combination in dealing with this function. Our results support the hypothesis that a well-focused micro population is more accurate and efficient than existing techniques, representing (that of micro-algorithms) a serious competitor because of its efficiency and accuracy. In fact, the proposed (but never solved) HGBat function can be dealt with, showing the scalability and potential future uses of our technique. INDEX TERMS Highly difficult problems, hybrid algorithms, HappyCat, HGBat, RayES, micro-algorithms. M. OLGUÍN-CARBAJAL received the degree in communications and electronics engineering from Culhuacan Unit, Higher School of Mechanical and Electrical Engineering (ESIME), National Polytechnic Institute, México, in 1995, the master's degree in computer engineering, in 2001, and D.Sc. degree from the Center for Computer Research (CIC), in 2011. In 2013, he did the Postdoctoral Residency with the University of Málaga, Spain. He has been a Professor and a full-time Researcher with the Centro de Innovación y Desarrollo Tecnológico en Cómputo (CIDETEC), since 2005. He has published several articles related to intelligent computing, computer graphics, virtual reality, and electronic applications in mobile devices.

No abstract

This paper presents a novel micro-segmented genetic algorithm (μsGA) to identify the best solution for the locomotion of a quadruped robot designed on a rectangular ABS plastic platform. We compare our algorithm with three similar algorithms found in the specialized literature: a standard genetic algorithm (GA), a micro-genetic algorithm (μGA), and a micro artificial immune system (μAIS). The quadruped robot prototype guarantees the same conditions for each test. The platform was developed using 3D printing for the structure and can accommodate the mechanisms, sensors, servomechanisms as actuators. It also has an internal battery and a multicore embedded system (mES) to process and control the robot locomotion. This research proposes a μsGA that segments the individual into specific bytes. μGA techniques are applied to each segment to reduce the processing time; the same benefits as the GA are obtained, while the use of a computer and the high computational resources characteristic of the GA are avoided. This is the reason why some research in robot locomotion is limited to simulation. The results show that the performance of μsGA is better than the three other algorithms (GA, μGA and AIS). The processing time was reduced using a mES architecture that enables parallel processing, meaning that the requirements for resources and memory were reduced. This research solves the problem of continuous locomotion of a quadruped robot, and gives a feasible solution with real performance parameters using a μsGA bio-micro algorithm and a mES architecture.

The calculation of Normalized Difference Vegetation Index (NDVI) has been studied in literature by multiple authors inside the remote sensing field and image processing field, however its application in large image files as satellite images restricts its use or need preprocessed phases to compensate for the large amount of resources needed or the processing time. This paper shown the implementation strategy to calculates NDVI for satellite images in RAW format, using the benefits of economic Supercomputing that were obtained by the video cards or Graphics Processing Units (GPU). Our algorithm outperforms other works developed in NVIDIA CUDA, the images used were provided by NASA and taken by Landsat 71 located on the Mexican coast, Ciudad del Carmen, Campeche.

Deep learning is a branch of machine learning and this technique allows us to create classifiers. We must find the best dataset size for a classifier process to permit using less time and give good accuracy. In this paper we will propose models with different deep layers and size dimensions for detecting the best model to solve a task that needs quick time processing.