A novel Decentralized Noisy Model Update Tracking Federated Learning algorithm (FedNMUT) is proposed that is tailored to function efficiently in the presence of noisy communication channels that reflect imperfect information exchange. This algorithm uses gradient tracking to minimize the impact of data heterogeneity while minimizing communication overhead. The proposed algorithm incorporates noise into its parameters to mimic the conditions of noisy communication channels, thereby enabling consensus among clients through a communication graph topology in such challenging environments. FedNMUT prioritizes parameter sharing and noise incorporation to increase the resilience of decentralized learning systems against noisy communications. Theoretical results for the smooth non-convex objective function are provided by us, and it is shown that the ϵ−stationary solution is achieved by our algorithm at the rate of $O\left( {\frac{1}{{\sqrt T }}} \right)$, where T is the total number of communication rounds. Additionally, via empirical validation, we demonstrated that the performance of FedNMUT is superior to the existing state-ofthe-art methods and conventional parameter-mixing approaches in dealing with imperfect information sharing. This proves the capability of the proposed algorithm to counteract the negative effects of communication noise in a decentralized learning framework.

Multimodal federated learning (FL) aims to enrich model training in FL settings where devices are collecting measurements across multiple modalities (e.g., sensors measuring pressure, motion, and other types of data). However, key challenges to multimodal FL remain unaddressed, particularly in heterogeneous network settings: (i) the set of modalities collected by each device will be diverse, and (ii) communication limitations prevent devices from uploading all their locally trained modality models to the server. In this paper, we propose Federated Multimodal Fusion learning with Selective modality communication (FedMFS), a new multimodal fusion FL methodology that can tackle the above mentioned challenges. The key idea is the introduction of a modality selection criterion for each device, which weighs (i) the impact of the modality, gauged by Shapley value analysis, against (ii) the modality model size as a gauge for communication overhead. This enables FedMFS to flexibly balance performance against communication costs, depending on resource constraints and application requirements. Experiments on the real-world ActionSense dataset demonstrate the ability of FedMFS to achieve comparable accuracy to several baselines while reducing the communication overhead by over 4x.

The problem of actuator fault estimation and fault-tolerant control for a class of uncertain nonlinear systems using Takagi-Sugeno fuzzy models is investigated. A design procedure for nonfragile proportional-integral (PI) observer is proposed to estimate the states of the nonlinear system and reconstruct the abrupt (modeled as step-like faults) and incipient fault signals. Subsequently, a nonfragile fault-tolerant controller is constructed, which is informed by the PI observer. Sufficient conditions of the existence of the PI observer and the fault-tolerant controller are provided in the form of linear matrix inequalities. The proposed fault-tolerant control architecture is tested on two numerical examples.

Variable neural adaptive robust control strategies are proposed for the output tracking control of a class of multiinput multioutput uncertain systems. The controllers incorporate a novel variable-structure radial basis function (RBF) network as the self-organizing approximator for unknown system dynamics. It can determine the network structure online dynamically by adding or removing RBFs according to the tracking performance. The structure variation is systematically considered in the stability analysis of the closed-loop system using a switched system approach with the piecewise quadratic Lyapunov function. The performance of the proposed variable neural adaptive robust controllers is illustrated with simulations.

In this paper, a generalized Brain-State-in-a-Box (gBSB)-based hybrid neural network is proposed for storing and retrieving pattern sequences. The hybrid network consists of autoassociative and heteroassociative parts. Then, a large-scale image storage and retrieval neural system is constructed using the gBSB-based hybrid neural network and the pattern decomposition concept. The notion of the deadbeat stability is employed to describe the stability property of the vertices of the hypercube to which the trajectories of the gBSB neural system are constrained. Extensive simulations of large scale pattern and image storing and retrieval are presented to illustrate the results obtained.

Real-time approximators for continuous-time dynamical systems with many inputs are presented. These approximators employ a novel self-organizing radial basis function (RBF) network, which varies its structure dynamically to keep the prescribed approximation accuracy. The RBFs can be added or removed online in order to achieve the appropriate network complexity for the real-time approximation of the dynamical systems and to maintain the overall computational efficiency. The performance of this variable structure RBF network approximator with both Gaussian RBF (GRBF) and raised-cosine RBF (RCRBF) is analyzed. The compact support of RCRBF enables faster training and easier output evaluation of the network than that of the network with GRBF. The proposed real-time self-organizing RBF network approximator is then employed to approximate both linear and nonlinear dynamical systems to illustrate the effectiveness of our proposed approximation scheme, especially for higher order dynamical systems. The uniform ultimate boundedness of the approximation error is proved using the second method of Lyapunov.

MOTIVATION: The still emerging combination of technologies that enable description and characterization of all expressed proteins in a biological system is known as proteomics. Although many separation and analysis technologies have been employed in proteomics, it remains a challenge to predict peptide behavior during separation processes. New informatics tools are needed to model the experimental analysis method that will allow scientists to predict peptide separation and assist with required data mining steps, such as protein identification. RESULTS: We developed a software package to predict the separation of peptides in strong anion exchange (SAX) chromatography using artificial neural network based pattern classification techniques. A multi-layer perceptron is used as a pattern classifier and it is designed with feature vectors extracted from the peptides so that the classification error is minimized. A genetic algorithm is employed to train the neural network. The developed system was tested using 14 protein digests, and the sensitivity analysis was carried out to investigate the significance of each feature. AVAILABILITY: The software and testing results can be downloaded from ftp://ftp.bbc.purdue.edu.

A class of interconnected neural networks composed of generalized Brain-State-in-a-Box (gBSB) neural subnetworks is considered. Interconnected gBSB neural network architectures are proposed along with their stability conditions. The design of the interconnected neural networks is reduced to the problem of solving linear matrix inequalities (LMIs) to determine the interconnection parameters. A method for solving LMIs is devised generating the solutions that, in general, are further away from zero than the corresponding solutions obtained using MATLAB's LMI toolbox, thus resulting in stronger interconnections between the subnetworks. The proposed architectures are then used to construct neural associative memories. Simulations are performed to illustrate the results obtained.

Fuzzy adaptive tracking controllers for a class of uncertain nonlinear dynamical systems are proposed and analyzed. The controllers consist of adaptive and robustifying components whose role is to nullify the effects of uncertainties and to achieve a desired tracking performance. The interactions between the two components have been investigated. The closed-loop system driven by the proposed controllers is shown to be stable with all the adaptation parameters being bounded. In particular, the proposed controllers guarantee uniform ultimate boundedness of the tracking error and the time bound of the uniform ultimate boundedness is obtained. An upper bound on the steady-state tracking error is obtained as a function of the gain of the robustifying term and the parameters of the adaptive component. The controllers are tested on an inverted pendulum and simulation results are included. A comparison of the proposed controllers with the ones in the literature is conducted.

This paper is concerned with large scale associative memory design. A serious problem with neural associative memories is the quadratic growth of the number of interconnections with the problem size. An overlapping decomposition algorithm is proposed to attack this problem. Specifically, a pattern to be processed is decomposed into overlapping sub-patterns. Then, neural sub-networks are constructed that process the sub-patterns. An error correction algorithm operates on the outputs of each sub-network in order to correct the mismatches between sub-patterns that are obtained from the independent recall processes of individual sub-networks. The performance of the proposed large scale associative memory is illustrated using two-dimensional images. It is shown that the proposed method reduces the computing cost of the design of the associative memories compared with non-interconnected associative memories.

A typical antilock brake system (ABS) senses when the wheel lockup is to occur, releases the brakes momentarily, and then reapplies the brakes when the wheel spins up again. In this paper, a genetic neural fuzzy ABS controller is proposed that consists of a nonderivative neural optimizer and fuzzy-logic components (FLCs). The nonderivative optimizer finds the optimal wheel slips that maximize the road adhesion coefficient. The optimal wheel slips are for the front and rear wheels. The inputs to the FLC are the optimal wheel slips obtained by the nonderivative optimizer. The fuzzy components then compute brake torques that force the actual wheel slips to track the optimal wheel slips; these torques minimize the vehicle stopping distance. The FLCs are tuned using a genetic algorithm. The performance of the proposed controller is compared with the case when maximal brake torques are applied causing a wheel lockup, and with the case when wheel slips are kept constant while the road surface changes.

Synchronous and asynchronous information system neural models are proposed that are hybrids of Pawlak's information system and Brain-State-in-a-Box (BSB) neural models. The stability of the proposed models is studied using LaSalle's Invariance Principle. Applications to an analysis of the United Nations activities are presented as examples.

A Lyapunov-based stabilizing control design method for uncertain nonlinear dynamical systems using fuzzy models is proposed. The controller is constructed using a design model of the dynamical process to be controlled. The design model is obtained from the truth model using a fuzzy modeling approach. The truth model represents a detailed description of the process dynamics. The truth model is used in a simulation experiment to evaluate the performance of the controller design. A method for generating local models that constitute the design model is proposed. Sufficient conditions for stability and stabilizability of fuzzy models using fuzzy state feedback controllers are given. The results obtained are illustrated with a numerical example involving a four-dimensional nonlinear model of a stick balancer.

A Lyapunov-based approach is used to derive a sufficiency condition for stabilizing a class of fuzzy system models using linear controllers. A design algorithm for constructing stabilizing linear controllers is given. The results obtained are illustrated with a design of a linear stabilizing controller for a system consisting of an inverted pendulum mounted on a cart. The controller is designed using a fuzzy model of the system and tested on the original system model.

Continuous-time neural networks for solving convex nonlinear unconstrained programming problems without using gradient information of the objective function are proposed and analyzed. Thus, the proposed networks are nonderivative optimizers. First, networks for optimizing objective functions of one variable are discussed. Then, an existing one-dimensional optimizer is analyzed, and a new line search optimizer is proposed. It is shown that the proposed optimizer network is robust in the sense that it has disturbance rejection property. The network can be implemented easily in hardware using standard circuit elements. The one-dimensional net is used as a building block in multidimensional networks for optimizing objective functions of several variables. The multidimensional nets implement a continuous version of the coordinate descent method.

The design problem of generalized brain-state-in-a-box (GBSB) type associative memories is formulated as a constrained optimization program, and "designer" neural networks for solving the program in real time are proposed. The stability of the designer networks is analyzed using Barbalat's lemma. The analyzed and synthesized neural associative memories do not require symmetric weight matrices. Two types of the GBSB-based associative memories are analyzed, one when the network trajectories are constrained to reside in the hypercube [-1, 1](n) and the other type when the network trajectories are confined to stay in the hypercube [0, 1](n). Numerical examples and simulations are presented to illustrate the results obtained.

In this paper we study three different classes of neural network models for solving linear programming problems. We investigate the following characteristics of each model: model complexity, complexity of individual neurons, and accuracy of solutions. Simulation examples are given to illustrate the dynamical behavior of each model.

We analyze the convergence properties of the Widrow-Hoff delta rule applied to McCulloch-Pitts type neurons. We give sufficiency conditions under which the learning parameters converge and conditions under which the learning parameters diverge. In particular, we analyze how the learning rate affects the convergence of the learning parameters.

Presents a novel synthesis procedure to realize an associative memory using the Generalized-Brain-State-in-a-Box (GBSB) neural model. The implementation yields an interconnection structure that guarantees that the desired memory patterns are stored as asymptotically stable equilibrium points and that possesses very few spurious states. Furthermore, the interconnection structure is in general non-symmetric. Simulation examples are given to illustrate the effectiveness of the proposed synthesis method. The results obtained for the GBSB model are successfully applied to other neural network models.

Deals with the use of neural networks to solve linear and nonlinear programming problems. The dynamics of these networks are analyzed. In particular, the dynamics of the canonical nonlinear programming circuit are analyzed. The circuit is shown to be a gradient system that seeks to minimize an unconstrained energy function that can be viewed as a penalty method approximation of the original problem. Next, the implementations that correspond to the dynamical canonical nonlinear programming circuit are examined. It is shown that the energy function that the system seeks to minimize is different than that of the canonical circuit, due to the saturation limits of op-amps in the circuit. It is also noted that this difference can cause the circuit to converge to a different state than the dynamical canonical circuit. To remedy this problem, a new circuit implementation is proposed.

A stability analysis is performed for the brain-state-in-a-box (BSB) neural models with weight matrices that need not be symmetric. The implementation of associative memories using the analyzed class of neural models is also addressed. In particular, the authors modify the BSB model so that they can better control the extent of the domains of attraction of stored patterns. Generalizations of the results obtained for the BSB models to a class of cellular neural networks are also discussed.

The problem of robust stability and convergence of learning parameters of adaptation algorithms in a noisy environment for the single preceptron is addressed. The case in which the same input pattern is presented in the adaptation cycle is analyzed. The algorithm proposed is of the Widrow-Hoff type. It is concluded that this algorithm is robust. However, the weight vectors do not necessarily converge in the presence of measurement noise. A modified version of this algorithm in which the reduction factors are allowed to vary with time is proposed, and it is shown that this algorithm is robust and that the weight vectors converge in the presence of bounded noise. Only deterministic-type arguments are used in the analysis. An ultimate bound on the error in terms of a convex combination of the initial error and the bound on the noise is obtained.

The authors propose a new class of adaptation algorithms for single- and multilayer perceptrons with discontinuous nonlinearities. The behavior of the proposed algorithms is shown on an application example and simulation results are included. The simulations were performed using the SIMNON package developed for purpose of simulation of nonlinear systems. The results can be used to control unknown dynamic systems using neural controllers. Indeed, many robust control algorithms utilize the inverse dynamics of the plant to be controlled. Thus, the proposed structures where the perceptrons are the inverse system model identifiers should constitute a part of the controller.>

The performance of controllers for a class of uncertain dynamical systems is analyzed in the presence of unmodeled actuator and sensor dynamics. It is shown that in uncertain systems with variable-structure control that contain small unmodeled inertias the sliding mode may be retained if an observer is utilized. The influence of the variable-structure observer on the performance of some control strategies used in the control of uncertain systems is also investigated. Finally, simulation results for different control strategies applied to a linearized model of a two-link manipulator subject to an uncertainty and the effects of a sensor and observer are presented.>

The design of variable-structure control (VSC) systems for a class of multivariable, nonlinear, time-varying systems is presented. Using the Utkin-Drazenovic method of equivalent control and generalized Lyapunov stability concepts, the VSC design is described in a unified manner. Complications that arise due to multiple inputs are examined, and several approaches useful in overcoming them are developed. Recent developments are investigated, as is the kinship of VSC and the deterministic approach to the control of uncertain systems. All points are illustrated by numerical examples. The recent literature on VSC applications is surveyed.>

The problem of estimating the region of stability for uncertain systems with bounded controllers and the sliding mode requirement is examined. A novel type of controller for a class of linear time-invariant systems subject to uncertainties is proposed. A transformation for decoupling the fast and slow states is utilized to investigate the stability-domain estimates of the system. The results are then applied to a two-joint planar manipulator and illustrated by computer simulation. >

Control of a class of nonlinear/uncertain systems is discussed using a variable-structure systems approach. Observations of the states of such systems is also considered. The natural extension to an observer-controller design is illustrated using a computer simulation example of a theta -r manipulator. Next, the problem of path planning is addressed using a combined observer-controller strategy. The aspects of hardware implementation of the proposed observer-controller are then analyzed.>