Internet of things (IoT) has opened new horizons in connecting all sorts of devices to the internet.However, continuous demand for connectivity increases the cybersecurity risks, rendering IoT devices more prone to cyberattacks.At the same time, rapid advances in Deep Learning (DL)-based algorithms provide state-of-the-art results in many classification tasks, including classification of network traffic or system logs.That said, deep learning algorithms are considered computationally expensive as they require substantial processing and storage capacity.Sadly, IoT devices have limited resources, making renowned DL models hard to implement in this environment.In this paper we present a Residual Neural Network inspired DLbased Intrusion Detection System (IDS) that incorporates weight pruning to make the model more compact in size and resource consumption.Additionally, the proposed system leverages feature selection algorithms to reduce the feature-space size.The model was trained on the NSL-KDD dataset benchmark.Experimental results show that the proposed system is effective, being able to classify network traffic with an F1 score of up to 98.9% before the pruning and an F1 score of up to 97.5% after pruning 90% of network weights.

Ant Colony Optimization (ACO) is an acclaimed method for solving combinatorial problems proposed by Marco Dorigo in 1992 and has since been enhanced and hybridized many times. This paper proposes a novel modification of the algorithm, based on the introduction of a two-dimensional pheromone into a single-criteria ACO. The complex structure of the pheromone is supposed to increase ants’ awareness when choosing the next edge of the graph, helping them achieve better results than in the original algorithm. The proposed modification is general and thus can be applied to any ACO-type algorithm. We show the results based on a representative instance of TSPLIB and discuss them in order to support our claims regarding the efficiency and efficacy of the proposed approach.

Metaheuristics are universal optimization algorithms which should be used for solving difficult problems, unsolvable by classic approaches. In this paper we aim at constructing novel socio-cognitive metaheuristic based on castes, and apply several versions of this algorithm to optimization of time-delay system model. Besides giving the background and the details of the proposed algorithms we apply them to optimization of selected variants of the problem and discuss the results.

Following the introduction of the socio-cognitive caste-based algorithms into the classic evolutionary metaheuristics, in this paper we focus on similar task regarding agent-based universal optimization methods. We tackle EMAS and DE algorithms and enrich them also with TOPSIS-inspired mechanism. Besides giving the details of the methods and the background, we present preliminary results after applying those techniques to solving the problem of optimization of time-delay system model.

Abstract Different hybrid optimization metaheuristics (see the works of Talbi for classification) either assume the embedding of one algorithm (usually a metaheuristic) in another (for instance, a local search inside an evolutionary algorithm—a memetic algorithm) or creating a chain of algorithms. In this paper, such a chain combination of two algorithms (namely, the Ant Colony Optimization and Evolutionary Algorithm) is presented. However, because of the intrinsic differences between the two algorithms (a vector of labels and a pheromone table when solving the traveling salesman problem, for example), several dedicated algorithms for translating the solutions between these two representations of the problem are proposed. The hybrid algorithm constructed with the application of the translation methods turns out to be significantly better in solving the TSP compared to non‐hybrid versions (relevant experimental results are presented and discussed). This paves the way for new possibilities of constructing hybrid metaheuristics by putting together completely different ones (using different representations); the impact of the presented research is aimed far beyond the hybridization of only ant colony optimization and evolutionary algorithm.

In the paper an application of selected agent-based evolutionary computing models, such as flock-based multi agent system (FLOCK) and evolutionary multi-agent system (EMAS), to the problem of continuous optimisation is presented. It turns out, that hybridizing of agent-based paradigm with evolutionary computation brings a new quality to the meta-heuristic field, easily enhancing static individuals with possibilities of perception and interaction with other agents. The examination of selected benchmarks leads to the observation regarding the overall efficiency of the systems in comparison to the standard genetic algorithm (as defined by Michalewicz) and memetic versions of all the systems. The experiments confirm that the efficiency is dependent on the problem, however, the observed number of fitness function calls makes EMAS dominate over its competitors. This feature makes EMAS a promising solution for the problems with complex fitness functions, (such as inverse problems).

The new methods of diagnosis are often more sensitive and speed up its process allowing for a more effective treatment. Yet they need to be carefully tested before they can be used in practice. The paper concerns the problems of developing nanorobots, which would circulate in the bloodstream of the human body gathering information about its condition. Simulation of these new ways of diagnosis is the main motivation behind the presented work. Introduction of such nanorobots into the bloodstream puts several requirements on their mechanical properties. Thus, our first goal is to build a model to simulate the blood flow with nanorobots present in the capillary vessels and medium-sized blood vessels. We split it into two separate initial models and implementations. The first one is a simulation of the blood flow using particlebased methods in order to determine the appropriate mechanical parameters of nanorobots and verify their behavior. The second one is a multi-agent simulation that will allow to evaluate the usefulness of the data collected and prototype the system performing functions of nanorobots within the specified constraints.

Existing solutions for agent-based systems turn out to be limited in some applications, like agent-based computing or simulations, where very large numbers of clearly defined agents interact heavily within a closed system. In those cases, fully-fledged, FIPA1 compliant environment introduce unnecessary overhead, but simple tools fail to scale when confronted to bigger problems. In this paper, we introduce an alternative agent environment called AgE, targeted at medium-sized simulation and computational applications, which use multi-agent and computational intelligence paradigms, but does not need full FIPA compliancy, and would benefit from a component-based approach and distributed computing capabilities. After giving a short review of selected popular multi-agent platforms, the main features of AgE are presented. Next, some basic usability topics are addressed. Then the most interesting architectural aspects of the platform are discussed. Finally, AgE possibilities are demonstrated with two example applications. Keywords— agent-based computing, component-based systems, agent-based simulation

In the paper an application of hybridized Evolutionary Multi-Agent System (EMAS) with local search (in memetic style) to the problem of continuous optimisation is presented. Before, the concept of evolutionary and memetic agent-based computing is given, the former being a computing paradigm researched for over 15 years, the latter being introduced recently. Two ways of memetic hybridization (Lamarckian and Baldwinian) are discussed, and examined in the course of experiments. In the presented experiments, evolutionary and memetic multi-agent systems are compared with classical evolutionary algorithm (Michalewicz model) implemented with allopatric speciation (island-model of evolutionary algorithm), based on a selected popular benchmark continuous optimization functions.

The paper presents an idea of agent-based meta-heuristic integrating a computational optimization system (evolutionary multi-agent system) with ant colony optimization technique. In the proposed model, chosen parameters of ant colonies may be encoded as genotypes and subjected to evolution process carried out by agents. The goal of the whole system is to search for the best solution of the discrete optimization problem based on the results of the ant colonies run using different parameters. The proposed concept forms a base for further research on bringing different interactions known in ant-colony optimization to the inter-agent level. The considerations are illustrated with preliminary experimental results obtained for parallel ant system solving quadratic assignment problem.

Agent-based framework dedicated to acquiring and processing heterogeneous data, collected in various Internet sources is presented. It is built upon a hierarchical, distributed computation system Age that has already been successfully used for various optimization and classification task.

This work presents a new evolutionary approach to searching for a global solution (in the Pareto sense) to a multiobjective optimisation problem. The novelty of the method proposed consists in the application of an evolutionary multi-agent system (EMAS) instead of classical evolutionary algorithms. Decentralisation of the evolution process in EMAS allows for intensive exploration of the search space, and the introduced mechanism of crowd allows for effective approximation of the whole Pareto frontier. In the paper the technique is described as well as preliminary experimental results are reported.