The automatic fitting of spiking neuron models to experimental data is a challenging problem. The integrate and fire model and Hodgkin–Huxley (HH) models represent the two complexity extremes of spiking neural models. Between these two extremes lies two and three differential-equation-based models. In this work, we investigate the problem of parameter estimation of two simple neuron models with a sharp reset in order to fit the spike timing of electro-physiological recordings based on two problem formulations. Five optimization algorithms are investigated; three of them have not been used to

In order to solve combinatorial optimization problem are used mainly hybrid heuristics. Inspired from nature, both genetic and ant colony algorithms could be used in a hybrid model by using their benefits. The paper introduces a new model of Ant Colony Optimization using multiple colonies with different level of sensitivity to the ant’s pheromone. The colonies react different to the changing environment, based on their level of sensitivity and thus the exploration of the solution space is extended. Several discussion follows about the fuzziness degree of sensitivity and its influence on the

Federated learning (FL) has recently been proposed as an emerging paradigm to build machine learning models using distributed training datasets that are locally stored and maintained on different devices in 5G networks while providing privacy preservation for participants. In FL, the central aggregator accumulates local updates uploaded by participants to update a global model. However, there are two critical security threats: poisoning and membership inference attacks. These attacks may be carried out by malicious or unreliable participants, resulting in the construction failure of global

This paper deals with solving large instances of the Linear Sum Assignment Problems (LSAPs) under realtime constraints, using Graphical Processing Units (GPUs). The motivating scenario is an industrial application for P2P live streaming that is moderated by a central tracker that is periodically solving LSAP instances to optimize the connectivity of thousands of peers. However, our findings are generic enough to be applied in other contexts. Our main contribution is a parallel version of a heuristic algorithm called Deep Greedy Switching (DGS) on GPUs using the CUDA programming language. DGS

The radical change in technology and high customer expectations and demand are pushing the service desk to provide a competent service for the customers. And this service should be faster, more flexible and gives a more accurate response and high user experience. Replacing the human call agent’s activities with a cognitive agent could improve the user experience by providing a faster response to customer’s requests, removing the human errors and giving 24/7 support with less operational cost. The cognitive agent has the potential to scale personalized and tailored interactions. And this will

Being sarcastic is to say something and to mean something else. Detecting sarcasm is key for social media analysis to differentiate between the two opposite polarities that an utterance may convey. Different techniques for detecting sarcasm are varying from rule-based models to Machine Learning and Deep Learning models. However, researchers tend to leverage Deep Learning in detecting sarcasm recently. On the other hand, the Arabic language has not witnessed much improvement in this research area. Bridging the gap in sarcasm detection of the Arabic language is the target behind this work. In

Using generic (non)extensive statistics, in which the underlying system likely autonomously manifests its extensive and nonextensive statistical nature, we extract various fit parameters from the CMS experiment and compare these to the corresponding results obtained from Tsallis and Boltzmann statistics. The present study is designed to indicate the possible variations between the three types of statistical approaches and characterizes their dependence on collision energy, multiplicity, and size of the system of interest. We analyze the transverse momentum spectra pT of the strange hadrons Ks

For blind and visually impaired (BVI) people, it is an arduous task to recognize clothing patterns and colors. It is especially complex to recognize them automatically. This is a highly researched area in image processing. This paper provides BVI with the ability to detect patterns and colors without depending on personal assistance, leading to increasing their confidence. The user first captures an image; then, our system detects colors and patterns. The color model used here is the hue saturation value (HSV). The main colors detected by our system are magenta, cyan, blue, green, red and

Established Internet of Things (IoT) platforms suffer from their inability to determine whether an IoT app is secure or not. A security analysis system (SAS) is a protective shield against any attack that breaks down data privacy and security. Its main task focuses on detecting malware and verifying app behavior. There are many SASs implemented in various IoT applications. Most of them build on utilizing static or dynamic analysis separately. However, the hybrid analysis is the best for obtaining accurate results. The SAS provides an effective outcome according to many criteria related to the

With the web-scale data volumes and high velocity of generation rates, it has become crucial that the training process for recommender systems be a continuous process which is performed on live data, i.e., on data streams. In practice, such systems have to address three main requirements including the ability to adapt their trained model with each incoming data element, the ability to handle concept drifts and the ability to scale with the volume of the data. In principle, matrix factorization is one of the popular approaches to train a recommender model. Stochastic Gradient Descent (SGD) has