Philippine Computing Journal.

Includes index and bibliographical references.

Spiking Neural P Systems: Balance and Homogeneity -- Evolving Spiking Neural P Systems with Polarization -- Evolving Spiking Neural P Systems by Fixing Neurons, and Varying Rules and Synapses -- A Short Survey of Stochastic Computing Models -- Periodic Dynamical Behavior in SN P Systems and the Presence of Cycles in their Graph Representation.

[Article Title: Spiking Neural dP Systems: Balance and Homogeneity / Kelvin C. Buño, Francis George C. Cabarle, and Jherico Gabriel Q. Torres, p.1-10] Abstract: This work explores some properties of Spiking Neural PSystems, namely: (1) balance and, (2) homogeneity. This class of Spiking Neural P Systems uses Extended SNP systems with Request Rules. As a case study, this work presents two SNdP systems that accept the language Lww={ww|w∈{b1,···,bk}n,n≥1}, where these SNdP systems differ in the balance of the input partition, the homogeneity of components of the system, or the number of SNP components. An analysis of the communication cost and running time is given for the two presented SNdP system.;[Article Title: Evolving Spiking Neural P Systems with Polarization / Jules Gerard E. Juico, Jerico L. Silapan, Francis George C. Cabarle, Ivan Cedric Macababayao, and Ren Tristan De la Cruz, p.11-20] Abstract: In this work, we introduce a representation of spikingneural P systems with polarization (or PSNP Systems)and an algorithm for simulation. An existing represen-tation and algorithm for spiking neural P systems ismodified to handle neurons with polarizations, neuronsfiring spikes and charges, and rules are checked usingthe charge of the neuron instead of regular expressions.In addition to the representation and algorithm, thiswork also presents a genetic algorithm framework (GAframework) that aims to reduce the number of resources(rules and synapses) of an existing PSNP system. We usethe framework on PSN P systems that perform bitwiseAND and OR. The GA framework will have two selectionmethods, Fitness Proportionate and Tournament Selec-tion. A discussion on the effectiveness of the frameworkin obtaining a PSN P System with less number of rulesand synapses will be done at the end. ;[Article Title: Evolving Spiking Neural P Systems by Fixing Neurons, and Varying Rules and Synapses / Carlo Cezar R. Zarate, Francis George C. Cabarle, Ivan Cedric Macababayao, and Ren Tristan Dela Cruz, p.21-30] Abstract: In this study, we explore evolving Spiking Neural PSystems, where a genetic algorithm framework is applied on Spiking Neural P Systems. The framework aims to (1)reduce the number of rules of an SN P system with 100%fitness and to (2) find a 100% fit SN P system using an initial SN P system with less than 100% fitness. Fitness, in this paper, is a measurement from 0% to 100% of an SN P system's accuracy in performing its intended function by getting the longest common substring of its output and its ideal output. The framework is limited to modifying the synapses and the rules of the given SNP systems. The framework is tested on bitwise addition and subtraction SN P systems, under three categories: Type A, with an initial fitness of 100%, Type B, with an initial fitness less than 100%, and Type C, with an initial fitness less than 100% fitness and with extra neurons. Such categories refer to the different characteristics of the SN P systems which will be the basis for initial population of our experiments.;[Article Title: A Short Survey of Stochastic Computing Models / Prometheus Peter L. Lazo, Francis George C. Cabarle, and Jan Michael C. Yap, p.31-38] Abstract: In this work, several stochastic computing models are investigated on how its stochastic process was introduced. While not all existing stochastic computing models are investigated in this work, most models included in this study involve a stochastic selection process with different approaches on the application of random variables and probabilities. Some of the results of this short survey were used for the recently introduced model known as spiking Neural P systems with stochastic application of rules (⋆SN P) during the 2020 Inter-national (electronic) Conference on Membrane Computing.;[Article Title: Periodic Dynamical Behavior in SN P Systems and the Presence of Cycles in their Graph Representation / Gyenn Neil Ibo and Henry N. Adorna, p.39-43] Abstract: In this paper, we explored how the topology of a genera-tive Spiking Neural P system (SN P system) relates withits dynamic behavior. Specifically, we proved our claim thatperiodicity in an SN P system implies presence of cycles inthe graph representing the system (where neurons are rep-resented as nodes and synapses are rpresented as directededges). Moreover, we extend the notion of periodicity from deterministic , one-rule-per-neuron generative SN P systemsto the non-deterministic, multi-rules-per-neuron generativeSN P systems. We also emphasize the importance of periodicity in generative SN P systems, and proved that a generative SN P system can generate an infinite number of outputs ifand only if it is periodic