Leszek Rutkowski

• Częstochowa University of Technology

This paper proposes a novel prescribed performance control scheme for stochastic micro-electro-mechanical system (MEMS) gyroscopes, addressing three critical issues overlooked by existing methods: control torque oscillation during rapid convergence, deviation in steady-state tracking errors in a global asymmetric design, and violation of monotonic...

Metaheuristics, such as evolutionary algorithms (EAs) have been proven to be (also theoretically, see, for example, the works of Michael Vose [1]) universal optimization methods. Previous works (Zbigniew Skolicki and Kenneth De Jong [2]) investigated impact of migration intervals on island models of EA’s in their works. Here we explore different mi...

In longwall coal mining, high methane concentrations pose a challenge. Cyber-physical air and gas protection systems ensure operational safety. However, safeguarding methane sensors from unauthorized interference is a key issue. This research aims to develop an intelligent method for detecting disturbances in methane sensor operation, preventing mi...

Socio-cognitive computing is a paradigm developed for the last several years in our research group. It consists of introducing mechanisms inspired by inter-individual learning and cognition into metaheuristics. Different versions of the paradigm have been successfully applied in hybridizing Ant Colony Optimization (ACO), Particle Swarm Optimization...

In this paper, the scale reduction problem of continuous-time Markov chains (CT-MCs) and continuous-time controlled Markov chains (CT-CMCs) are disserted both from the bisimulation perspective. Based on the features of bisimulation, the reachability, macro-controllability, controllability, and stabilizability of CT-MCs and CT-CMCs, particularly, th...

In this paper, we investigate the leader-follower consensus of multi-agent systems over finite fields, which model agents with limited capacities for storing, processing, and transmitting the information, from the perspectives of the transition graph and the characteristic polynomial of the network matrix, respectively. By the features of dynamics...

This article focuses on the fuzzy $\mathcal {H}_{\infty }$ control problem for a class of semi-Markov jump singularly perturbed nonlinear systems with actuator saturation. Nonlinearities in the underlying systems are tackled with the Takagi–Sugeno fuzzy model. As distinct from the previous achievements, the case that the semi-Markov kernel with p...

One of the major challenges in modern artificial neural network training methods is reducing the learning time. To address this issue, a promising approach involves the continuous selection of the most crucial elements from the training set, utilizing data stream analysis. However, transitioning to this new learning paradigm raises several question...

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 p...

Dear Editor, This letter deals with the set stabilization of stochastic Boolean control networks (SBCNs) by the pinning control strategy, which is to realize the full control for systems by imposing control inputs on a fraction of agents. The pinned agents are determined based on the information on the network structure, rather than the whole state...

Gas composition and light mode of industrial greenhouses are some of the most determining factors in the process of growing vegetable crops in greenhouse conditions. The intelligentisation of information technologies for monitoring and control based on artificial intelligence methods can increase the efficiency of agrotechnical procedures for green...

This article presents a decentralized learning control method for a class of partially unknown nonlinear systems with asymmetric control input constraints and mismatched interconnections via a novel dynamic event-triggering condition. By employing an integral reinforcement learning strategy, the system drift dynamics can be avoided in the learning...

In this paper, observable stochastic graphs and detetectable stochastic graphs are, respectively, defined with the detailed implementation for the observability and detectability of stochastic discrete-time and discrete-state dynamic systems. More specifically, they are generally two classes of vertex-colored and edge-labeled graphs rendering a w...

Metaheuristic methods are designed to solve continuous and discrete problems. Such methods include population based algorithms (PBAs). They are distinguished by the flexibility of defining the fitness function, therefore they are a good alternative to gradient methods. However, creating new variants of PBAs that work similarly and differ in detail...

In this paper, the local consensus problem of nonlinear time-delay multi-agent systems with switching topologies via distributed saturated impulsive control is discussed and the maximum domain of attraction is well estimated. Specifically, we develop a new estimation approach that is quite distinct from the contractive invariant set to estimate the...

The dynamical system with a second-order+first-order structure has been proven to be effective for solving convex optimization problems with equality constraints. In this paper, this structure is extended to distributed optimization to deal with set-constrained convex optimization problems under strongly connected and weight-balanced directed topol...

This work proposes a fully distributed moving horizon estimation method with an event-triggered communication strategy over wireless sensor networks. The proposed method calculates a local state estimation of each sensor by minimizing a quadratic objective function, which involves a fused arrival cost that is computed in a distributed manner. This...

In this brief, we consider the stability of inertial memristor-based neural networks with time-varying delays. First, delayed inertial memristor-based neural networks are modeled as continuous systems in the flux-current-voltage-time domain via the mathematical model of Hewlett-Packard (HP) memristor. Then, they are reduced to delayed inertial neur...

This paper deals with global synchronization problem of multiple discrete-time Markovian jump memristor-based neural networks (DTMJMNNs) with mixed mode-dependent delays via a novel event-triggered impulsive coupling control (ETICC). The parameters of the multiple DTMJMNNs and the mixed time delays (both discrete and distributed delays) switch rand...

In this paper, we investigate the stabilization of Markovian jump Boolean control networks by a kind of event-triggered control, which is essentially an intermittent control scheme. Firstly, a novel condition for the stability of Markovian jump Boolean networks is obtained based on the recurrence of finite-state homogeneous Markov chains. After tha...

This paper considers the fixed-time distributed optimization problem with consensus constraint and strongly convex local cost functions, and a distributed optimization algorithm involving two stages is designed. The first stage is to make each agent converge to its own locally optimal state from any initial state in fixed time by designing distribu...

In this paper, we propose the Convolutional Restricted Boltzmann Machine (CRBM) as a tool for detecting concept drift in time-varying data streams. Recently, it was demonstrated that the Restricted Boltzmann Machine (RBM) can be successfully applied to this task. A properly learned RBM contains information about the data probability distribution. T...

One of the greatest challenges facing researchers of machine learning algorithms nowadays is the desire to minimize the training time of these algorithms. One of the most promising and unexplored structures of the neural network is the Restricted Boltzmann Machine. In this paper, we propose to use the BBTADD algorithm for RBM training. The performa...

In this paper, finite-field networks (FFNs) with time delays are investigated. We introduce the characteristic matrix polynomial for delayed FFNs and utilize it to analyze the dynamics considered in view of linear recursion theory. It is shown that delayed FFNs behave in a pattern similar to non-delayed FFNs in the sense that any state sequence fin...

This article focuses on the composite ${H}_{∞ }$ synchronization problem for jumping reaction-diffusion neural networks (NNs) with multiple kinds of disturbances. Due to the existence of disturbance effects, the performance of the aforementioned system would be degraded; therefore, improving the control performance of closed-loop NNs is the main...

This paper investigates the dynamical multisynchronization (DMS) and static multisynchronization (SMS) problems for a class of delayed coupled multistable memristive neural networks (DCMMNNs) via a novel hybrid controller which includes delayed impulsive control and state feedback control. Based on the state space partition method and the geometric...

In this brief, we consider the problem of descriptors construction for the task of content-based image retrieval using deep neural networks. The idea of neural codes, based on fully connected layers' activations, is extended by incorporating the information contained in convolutional layers. It is known that the total number of neurons in the convo...

In this article, we study the finite-time stabilization and the asymptotic stabilization with probability one of Markovian jump Boolean control networks (MJBCNs) by sampled-data state feedback controls (SDSFCs). Based on the semi-tensor product (STP), we introduce an augmented variable multiplied by the vector form of the switching signal and the s...

In this article, a periodic self-triggered impulsive (PSTI) control scheme is proposed to achieve synchronization of neural networks (NNs). Two kinds of impulsive gains with constant and random values are considered, and the corresponding synchronization criteria are obtained based on tools from impulsive control, event-driven control theory, and s...

The two-volume set LNAI 12854 and 12855 constitutes the refereed proceedings of the 20th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2021, held in Zakopane, Poland, in June 2021. Due to COVID 19, the conference was held virtually. The 89 full papers presented were carefully reviewed and selected from 195 submissio...

The two-volume set LNAI 12854 and 12855 constitutes the refereed proceedings of the 20th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2021, held in Zakopane, Poland, in June 2021. Due to COVID 19, the conference was held virtually. The 89 full papers presented were carefully reviewed and selected from 195 submissio...

This paper deals with the synchronization for discrete-time coupled neural networks (DTCNNs), in which stochastic perturbations and multiple delays are simultaneously involved. The multiple delays mean that both discrete time-varying delays and distributed delays are included. Time-triggered impulsive control (TTIC) is proposed to investigate the s...

A new type of asymptotic stability for nonlinear hybrid neutral stochastic systems with constant delays was investigated recently, where the criteria depended on the delays’ sizes. Unfortunately, developed theory so far is not sufficient to deal with challenging problems of the decay rate, time-varying delays, and nonautonomous issues. These proble...

This article studies the constrained optimization problems in the quaternion regime via a distributed fashion. We begin with presenting some differences for the generalized gradient between the real and quaternion domains. Then, an algorithm for the considered optimization problem is given, by which the desired optimization problem is transformed i...

In this paper, we propose a method for rules generation for images, which can be further used to enhance the classification accuracy of a convolutional neural network as well as to increase the level of explainability of neural network decisions. For each image from the training set, a descriptor is created based on the information contained in the...

This paper presents a novel recommendation system for investment managers using real data from asset management companies. The recommender can be viewed as a fuzzy expert system. As a matter of fact, this is an explainable recommender that works as a one-class classifier with an explanation. The inference rules, explanations, and visualizations of...

In this paper, the problem of concept drift detection in data stream mining algorithms is considered. The autoencoder is proposed to be applied as a drift detector. The autoencoders are neural networks that are learned how to reconstruct input data. As a side effect, they are able to learn compact nonlinear codes, which summarize the most important...

Although recurrent neural networks (RNNs) perfectly solve many difficult problems, their computational complexity significantly increases training time. Therefore, the primary problem with applying RNNs is to shorten the time needed to train and operate a network. An effective solution to this problem is to use parallel processing. In the paper, a...

This article considers global exponential synchronization almost surely (GES a.s.) for a class of switched discrete-time neural networks (DTNNs). The considered system switches from one mode to another according to transition probability (TP) and evolves with mode-dependent average dwell time (MDADT), i.e., TP-based MDADT switching, which is more p...

In this paper, we consider the problem of descriptors construction for the task of content-based image retrieval using deep neural networks. The idea of neural codes, based on fully connected layers activations, is extended by incorporating the information contained in convolutional layers. It is known that the total number of neurons in the convol...

This paper investigates the set stabilization of impulsive Boolean control networks (IBCNs) by sampled-data state feedback control (SDSFC) based on a hybrid index model. It is worth mentioning that 2-D index model has the ability to characterize the instantaneousness of ideal impulses and describe complicated impulsive behaviors. To avoid Zeno phen...

This article addresses the investigation of sliding-mode control (SMC) for slow-sampling singularly perturbed systems (SPSs) with Markov jump parameters. As a new attempt, the SMC strategy is considered in the study of discrete-time Markov jump SPSs. Subsequently, in order to design a sliding-mode controller to ensure the stability of the proposed...

In this paper, we investigate the asymptotic output tracking control problem of probabilistic Boolean control networks. Firstly, based on the largest control invariant subset and the limit theory of Markov chains, the conditions are proposed to judge whether the output trajectory of a probabilistic Boolean control network can asymptotically track a...

In the literature, the effects of switching with average dwell time (ADT), Markovian switching, and intermittent coupling on stability and synchronization of dynamic systems have been extensively investigated. However, all of them are considered separately because it seems that the three kinds of switching are different from each other. This articl...

In this chapter, we will investigate the issue of automatic selection of ensemble components. The presented methodology allows guaranteeing, that a new component will be included into an ensemble only if it significantly improves the performance of the ensemble, not only for a current chunk of data but also for the whole stream. Additionally, the e...

A decision tree [1] is a data mining tool commonly used in data classification tasks. Apart from providing satisfactorily high accuracies, the results produced by decision trees are easily interpretable. A decision tree, in fact, divides attribute values space X into disjoint subspaces. The most common decision tree induction algorithms for static...

The problems of learning in non-stationary situations has rarely been a subject of studies even in a parametric case. Historically the first papers on learning in non-stationary environments where occasionally published in the sixties and seventies. The proper tool for solving such a type of problems seemed to be the dynamic stochastic approximatio...

One way of solving the problem of incompatibility between nonlinear split measures, like the information gain or the Gini gain, and the Hoeffding’s inequality is the application of another statistical tool, e.g. the McDiarmid’s inequality. Another way is to find a split measure which can be expressed as an arithmetic average of some random variable...

Since the Hoeffding’s inequality proved to be irrelevant in establishing splitting criteria for the information gain and the Gini gain, a new statistical tool has to be proposed. In this chapter, the McDiarmid’s inequality [1] is introduced, which is a generalization of the Hoeffding’s one to any nonlinear functions. Further extensions and analysis...

Data stream mining, as its name suggests, is connected with two basic fields of computer science, i.e. data mining and data streams. Data mining [1, 2, 3, 4] is an interdisciplinary subfield of computer science whose main aim is to develop tools and methods for exploring knowledge from large datasets. Data mining is strictly related to statistics,...

The literature concerning the supervised learning algorithms in data stream mining is dominated mainly by pattern classification methods. Only few of them deal with a non-stationary regression. Most of them rely on the Gaussian or Markov models, extend Support Vector Machine or Extreme Learning Machine to regression problems, implement regression t...

Among the data stream mining algorithms proposed so far in the literature most of them are devoted mainly to the data classification task [1, 2, 3]. Although there exist a lot of methods for classification of static datasets, they can hardly be adapted to deal with data streams. This is due to the features of the data stream such as potentially inf...

In this book, we studied the problem of data stream mining. Recently, it became a very important and challenging issue of computer science research. The reason is the enormous growth of data amounts generated in various areas of human activities. Data streams [1, 2, 3] are potentially of infinite size and often arrive at the system with very high r...

During constructing data stream algorithms the following three aspects have to be taken into consideration: accuracy, running time and required memory. However, in many cases, the fastest algorithms are less accurate than methods requiring high computational power and more time for data analysis. Therefore, to enhance the performance of the algorit...

In the previous chapters, various types of splitting criteria were proposed. Each of the presented criteria is constructed using one specific impurity measure (or, more precisely, the corresponding split measure function). Therefore we will refer to such criteria as ‘single’ splitting criteria. The experiments conducted in Chap. 5 demonstrate that...

In recent decades we are observing an exponential increase in the available digital data, generated in various areas of human activity. This growth is much faster with respect to the increase in the available processing capabilities. Apart from large volumes, the data produced by modern data sources are often dynamic and generated at very high rate...

Probabilistic neural networks (PNN), introduced by Specht [1, 2] have their predecessors in the theory of statistical pattern classification. In the fifties and sixties, problems of statistical pattern classification in the stationary case were accomplished by means of parametric methods, using the available apparatus of statistical mathematics (e....

Although ensembles of classifiers are one of the most popular tools to deal with data streams classification task [1, 2, 3, 4, 5], in the literature there is a lack of new approaches to creating ensembles of regression estimators [6, 7]. Most of the latest developments focus on the application of the regression estimators to solve very important re...

This book presents a unique approach to stream data mining. Unlike the vast majority of previous approaches, which are largely based on heuristics, it highlights methods and algorithms that are mathematically justified. First, it describes how to adapt static decision trees to accommodate data streams; in this regard, new splitting criteria are dev...

The two-volume set LNCS 12415 and 12416 constitutes the refereed proceedings of of the 19th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2020, held in Zakopane, Poland*, in October 2020. The 112 revised full papers presented were carefully reviewed and selected from 265 submissions. The papers included in the first...

This article investigates the output regulation problem of probabilistic $k$ -valued logical systems with delays by an intermittent control scheme. Two types of event-triggered control are designed via semi-tensor product (STP) of matrices. According to the algebraic state-space representation of probabilistic $k$ -valued systems with delays, t...

In the paper, we develop the mathematically justified stream data mining algorithm for solving regression problems. The algorithm is based on the Hermite expansions of drifting regression functions. The global convergence, in the \(L_2\) space, is proved both in probability and with probability one. The examples of several concept drifts to be hand...

This paper addresses the issue of data stream mining using the Restricted Boltzmann Machine (RBM). Recently, it was demonstrated that the RBM can be useful as a concept drift detector in data streams with time-changing probability density. In this paper, we consider another problem which often occurs in real-life data streams, i.e. incomplete data....

In this paper, we consider the problem of data stream mining with an application of the Restricted Boltzmann Machine (RBM). If the data incoming rate is very fast, an appropriate algorithm should be resource-aware and work as fast as possible. Two RBM learning algorithms are investigated, i.e. the Contrastive Divergence and the Persistent Contrasti...

In this paper, we present several approaches to configuration of deep convolutional neural networks for image classification. A common problem when creating deep structures is their proper designing and configuration. This paper shows the learning of the baseline model for image classification and its variations with different structures based on t...

This paper describes a method of pixel-level segmentation applied to parasite detection. Parasite diseases in most cases are detected by microscopic samples examination or by ELISA blood tests. The microscopic methods are less invasive and often used in veterinary, but they need more time to prepare and visually evaluate samples. Diagnosticians sea...

This paper proposes a quaternion-valued one-layer recurrent neural network approach to resolve constrained convex function optimization problems with quaternion variables. Leveraging the novel generalized Hamilton-real (GHR) calculus, the quaternion gradient-based optimization techniques are proposed to derive the optimization algorithms in the qua...

In this paper, we propose a recursive variant of the Parzen kernel density estimator (KDE) to track changes of dynamic density over data streams in a nonstationary environment. In stationary environments, well-established traditional KDE techniques have nice asymptotic properties. Their existing extensions to deal with stream data are mostly based...

In this paper, a novel procedure for regression analysis in the case of non-stationary data streams is presented. Despite numerous applications, the regression task is rarely considered in a scientific literature, e.g. compared to classification task. The proposed method applies an ensemble technique to deal with data streams (especially with conce...

Verification of the dynamic signature is an important issue of biometrics. There are many methods for the signature verification using dynamics of the signing process. Many of these methods are based on the so-called global features. In this paper we propose a new approach to the signature verification using global features. The proposed approach c...

Fuzzy systems are well suited for nonlinear modeling. They can be effectively used if their structure and structure parameters are properly chosen. Moreover, it should be ensured that system rules are clear and interpretable. In this paper we propose a new algorithm for automatic learning and new interpretability criteria of fuzzy systems. Interpre...

The two-volume set LNAI 10841 and LNAI 10842 constitutes the refereed proceedings of the 17th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2018, held in Zakopane, Poland in June 2018. The 140 revised full papers presented were carefully reviewed and selected from 242 submissions. The papers included in the first v...

The two-volume set LNAI 10841 and LNAI 10842 constitutes the refereed proceedings of the 17th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2018, held in Zakopane, Poland in June 2018. The 140 revised full papers presented were carefully reviewed and selected from 242 submissions. The papers included in the second...

One of the greatest challenges in data mining is related to processing and analysis of massive data streams. Contrary to traditional static data mining problems, data streams require that each element is processed only once, the amount of allocated memory is constant and the models incorporate changes of investigated streams. A vast majority of ava...

In this paper a method for nonparametric regression estimation in a time-varying environment is presented. The orthogonal series-based kernels are used to design learning procedures tracking non-stationary systems changes under non-stationary noise. The presented procedures, constructed in the spirit of generalized regression neural networks, are a...

In this paper the regression function methods based on Parzen kernels are investigated. Both the modeled function and the variance of noise are assumed to be time-varying. The commonly known kernel estimator is extended by adopting two popular tools often applied in concept drifting data stream scenario. The first tool is a sliding window, in which...

The most popular tools for stream data mining are based on decision trees. In previous 15 years, all designed methods, headed by the very fast decision tree algorithm, relayed on Hoeffding's inequality and hundreds of researchers followed this scheme. Recently, we have demonstrated that although the Hoeffding decision trees are an effective tool fo...

In this paper we propose a new approach for designing an ensemble applied to stream data classification. Our approach is supported by two theorems showing how to decide whether a new component should be added to the ensemble or not, based on the assumption that such an action should increase the accuracy of the ensemble not only for the current por...

Identity verification based on authenticity assessment of a handwritten signature is an important issue in biometrics. There are many effective methods for signature verification taking into account dynamics of a signing process. Methods based on partitioning take a very important place among them. In this paper we propose a new approach to signatu...

This paper presents a novel approach to visual objects classification based on generating simple fuzzy classifiers using local image features to distinguish between one known class and other classes. Boosting meta learning is used to find the most representative local features. The proposed approach is tested on a state-of-the-art image dataset and...