GPU implementation of evolving spiking neural P systems

Article Computer Science, Information Systems

A new P-Lingua toolkit for agile development in membrane computing

Ignacio Perez-Hurtado et al.

Summary: This paper presents the challenges of validating and testing P system models in membrane computing and proposes a new design of the P-Lingua language that allows users to define custom variants and derivation modes, among other improvements. It also includes a set of command-line tools for parsing and simulating P-Lingua 5 files.

INFORMATION SCIENCES (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Evolutionary Optimization of Spiking Neural P Systems for Remaining Useful Life Prediction

Leonardo Lucio Custode et al.

Summary: This study tackles the problem of RUL prediction for the first time using the SN P systems and proposes the use of a neuro-evolutionary algorithm to optimize the structure and parameters. The results show a reasonable compromise between performance and number of trainable parameters compared to complex deep networks.

ALGORITHMS (2022)

添加到收藏夹

Article Engineering, Chemical

Simulation of Spiking Neural P Systems with Sparse Matrix-Vector Operations

Miguel Angel Martinez-del-Amor et al.

Summary: This paper introduces a new simulation algorithm based on a novel compressed representation for sparse matrices, aiming to address the performance issues caused by sparse matrix representations in SNP systems. The study applies advanced concepts from parallel computing literature to improve the simulation efficiency for different variants of SNP systems.

PROCESSES (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Numerical Spiking Neural P Systems

Tingfang Wu et al.

Summary: A novel class of SN P systems, called numerical SN P (NSN P) systems, is proposed in this work, where information is encoded by the values of variables and described by continuous production functions. The computation power of NSN P systems is investigated, showing that they are Turing universal as number generating and accepting devices, even with simple linear production functions in each neuron. This suggests the potential for developing learning algorithms for NSN P systems.

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2021)

添加到收藏夹

Article Computer Science, Theory & Methods