Medina Huaman, Ollantay

Profile Picture

Filters

2007 - 200912010 - 20141
Reset filters

Settings

Citations

Publication Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Publication
    A neocortex-based model for pattern recognition
    (2014) Medina Huaman, Ollantay; Manian, Vidya; College of Engineering; Acuna, Edgar; Rodriguez, Domingo; Urintsev, Alexander; Department of Electrical and Computer Engineering; Velazquez, Esov
    The information society needs to process massive volumes of data in automated ways. The data is more meaningful if it is considered as sequences of patterns rather than isolated patterns. A trained human brain has outstanding capabilities for pattern recognition tasks, which could be very advantageous for automated pattern recognition systems. This thesis proposes a generic pattern recognition model. This model is mainly based on the known operation of the neocortex and is oriented to deal with sequences of patterns or data streams. The theoretical part of the model is formally stated and an implementation is outlined. The theoretical model also establishes a more general framework for treatment of space time data through a dimensionality reduction process. For a given instance of space time data, the process characterizes a space time region that might be called an invariant semantic representation. The model exhibits desirable properties for a pattern recognition system, such as spatial and temporal autoassociativity, spatial and temporal noise tolerance, recognition under sequence contextualization, and input prediction.
  • Thumbnail Image
    Publication
    Juegos dinámicos de persecución y evasión
    (2007) Medina Huaman, Ollantay; Vásquez Urbano, Pedro; College of Arts and Sciences - Sciences; Castillo, Paúl E.; Acar, Robert; Department of Mathematics; Rivera, Wilson
    Pursuit-Evasion games settle the foundations for Dynamic Games, a very active area in these days due to its great applicative potential. The Hamstrung Squad Car and the Homicidal Chaffeur are two zero-sum, two-player Pursuit-Evasion games with complete information, that this work studies in order to formulate and implement a numeric solution based on Dynamic Programming for this kind of games. The resulting algorithms are implemented using Matlab and have complexity of polynomial order, depending directly on the dimension of the problem and the number of directions that players can use. These algorithms can be reused to solve similar problems.