Robert Kozma

University of Memphis | U of M · Department of Mathematical Sciences previously Department of Computer Science

Abstract— Experimental studies using electrocorticograms (ECoGs) over the cortical surface indicate spatio-temporal dynamics in the form of amplitude modulation (AM) patterns, which intermittently collapse at theta rates and give rise to rapidly propagating phase modulated (PM) patterns. The observed dynamics has been shown to contain useful i...

Ametamodel establishes the constructs and rules governing the creation of models within a specific modeling methodology, such as model-based systems engineering (MBSE), which has gained increasing popularity among researchers and practitioners. The absence of a universally accepted metamodel can lead to various issues, which have not been adequatel...

It is of great interest to develop advanced sensory technologies allowing non-invasive monitoring of neural correlates of cognitive processing in people performing everyday tasks. A lot of progress has been reported in recent years in this research area using scalp EEG arrays, but the high level of noise in the electrode signals poses a lot of chal...

The new generation of digital intelligence technology enables knowledge creation, dissemination, and application to undergoing parallel changes. Scientific systems face an increasingly uncertain, diverse, and complex environment, making adopting multidisciplinary, interdisciplinary, and transdisciplinary approaches to research issues inevitable. Ex...

It has been proposed that meditative states show different brain dynamics than other more engaged states. It is known that when people sit with closed eyes instead of open eyes, they have different brain dynamics, which may be associated with a combination of deprived sensory input and more relaxed inner psychophysiological and cognitive states. He...

The integration of deep learning (DL) and fuzzy learning (FL) is considered a recently emerging and promising research direction in data embedding. The integrated fuzzy neural learning paradigm within deep learning-based recommendation helped improving the performance of both latent feature representation learning and multiple recommendation proble...

Decentralized science (DeSci) is a hot topic emerging with the development of Web3 or Web3.0 and decentralized autonomous organizations (DAOs) and operations. DeSci fundamentally differs from the centralized science (CeSci) and Open Science (OS) movement built in the centralized way with centralized protocols. It changes the basic structure and leg...

Deep Spiking Neural Networks (SNNs) with event-driven dynamics become increasingly popular in many challenging Machine Learning applications, based on their cheap and efficient computations. The discontinuity of the SNN dynamics, however, leads to problems in the learning process, resulting in performance loss, as the dominant gradient-based traini...

Spatio-temporal brain activity monitored by EEG recordings in humans and other mammals has identified beta/gamma oscillations (20–80 Hz), which are self-organized into spatio-temporal structures recurring at theta/alpha rates (4–12 Hz). These structures have statistically significant correlations with sensory stimuli and reinforcement contingencies...

Spiking neural networks (SNNs) have great potential for energy-efficient implementation of Deep Neural Networks (DNNs) on dedicated neuromorphic hardware. Recent studies demonstrated competitive performance of SNNs compared with DNNs on image classification tasks, including CIFAR-10 and ImageNet data. The present work focuses on using SNNs in combi...

In a previous study (Davis et al., 2018) we showed how meditative states significantly improve psychophysiological coherence when contrasted with baseline or other daily activities. However, meditation benefits can be compromised when individuals participate in energy consuming or unpleasant activities, which lead to stressful states generated as a...

This paper presents the first general set of results concerning a study we conducted in February - March 2015, as part of a body of researchers in 5 countries: The United States of America, United Kingdom, Lithuania, Saudi Arabia and New Zealand, led by the HeartMath Institute in California under the name, International Heart Rate Variability Synch...

This work studies the evolution of cortical networks during the transition from escape strategy to avoidance strategy in auditory discrimination learning in Mongolian gerbils trained by the well-established two-way active avoidance learning paradigm. The animals were implanted with electrode arrays centered on the surface of the primary auditory co...

Current Artificial Intelligence (AI) machine learning approaches perform well with similar sensors for data collection, training, and testing. The ability to learn and analyze data from multiple sources would enhance capabilities for Artificial Intelligence (AI) systems. This paper presents a deep learning-based multi-source self-correcting approac...

A 50th birthday is an important milestone in the life of any individual and certainly in the development of collectives and organizations. January 2021 is such a milestone in the life of the IEEE T ransactions of S ystems , M an , and C ybernetics , which had its very first issue published in January 1971. Our 50th Anniversary Issue celebrate...

To commemorate the 50th anniversary of the IEEE Transactions on Systems, Man, and Cybernetics: Systems, this article examines and reports on its past to current topical coverage of systems science and engineering toward exploring the evolving focus of the research community. Results of a systematic bibliometric analysis are presented with associate...

Spiking neural networks (SNNs) with a lattice architecture are introduced in this work, combining several desirable properties of SNNs and self-organized maps (SOMs). Networks are trained with biologically motivated, unsupervised learning rules to obtain a self-organized grid of filters via cooperative and competitive excitatory-inhibitory interact...

Non-invasive brain imaging techniques are popular tools for monitoring the cognitive state of human participants. This work builds on our previous studies using the HydroCel Geodesic Sensor Net, 256 electrodes dense-array electro-encephalography (EEG). The studies analyze dominant frequencies of temporal power spectral densities for each of the EEG...

Spiking neural networks (SNNs) have great potential for energy-efficient implementation of Deep Neural Networks (DNNs) on dedicated neuromorphic hardware. Recent studies demonstrated competitive performance of SNNs compared with DNNs on image classification tasks, including CIFAR-10 and ImageNet data. The present work focuses on using SNNs in combi...

Spiking neuron networks have been used successfully to solve simple reinforcement learning tasks with continuous action set applying learning rules based on spike-timing-dependent plasticity (STDP). However, most of these models cannot be applied to reinforcement learning tasks with discrete action set since they assume that the selected action is...

Leading mainstream image processing approaches produce excellent performance using convolutional neural networks trained by backpropagation (BP) learning rules. Unsu-pervised learning approaches have been popular due to their biological significance, though they typically underperform compared to BP results. In this work, we demonstrate that featur...

Neuroscientific theory suggests that dopaminergic neurons broadcast global reward prediction errors to large areas of the brain influencing the synaptic plasticity of the neurons in those regions. We build on this theory to propose a multi-agent learning framework with spiking neurons in the generalized linear model (GLM) formulation as agents, to...

Spiking neural networks (SNNs) are a promising candidate for biologically-inspired and energy efficient computation. However, their simulation is notoriously time consuming, and may be seen as a bottleneck in developing competitive training methods with potential deployment on neuromorphic hardware platforms. To address this issue, we provide an im...

Deep Reinforcement Learning (RL) demonstrates excellent performance on tasks that can be solved by trained policy. It plays a dominant role among cutting-edge machine learning approaches using multi-layer Neural networks (NNs). At the same time, Deep RL suffers from high sensitivity to noisy, incomplete, and misleading input data. Following biologi...

In recent years, spiking neural networks (SNNs) have demonstrated great success in completing various machine learning tasks. We introduce a method for learning image features with locally connected layers in SNNs using a spike-timing-dependent plasticity (STDP) rule. In our approach, sub-networks compete via inhibitory interactions to learn featur...

The mammalian olfactory bulb displays a prominent respiratory rhythm, linked to the sniff cycle and driven by sensory input from olfactory receptors in the nasal sensory epithelium. In rats and mice, respiratory frequencies occupy the same band as the hippocampal theta rhythm, which has been shown to be a key player in memory processes. Hippocampal...

Spiking neural networks (SNNs) with a lattice architecture are introduced in this work, combining several desirable properties of SNNs and self-organized maps (SOMs). Networks are trained with biologically motivated, unsupervised learning rules to obtain a self-organized grid of filters via cooperative and competitive excitatory-inhibitory interact...

In recent years, Spiking Neural Networks (SNNs) have demonstrated great successes in completing various Machine Learning tasks. We introduce a method for learning image features by \textit{locally connected layers} in SNNs using spike-timing-dependent plasticity (STDP) rule. In our approach, sub-networks compete via competitive inhibitory interacti...

Various implementations of Deep Reinforcement Learning (RL) demonstrated excellent performance on tasks that can be solved by trained policy, but they are not without drawbacks. Deep RL suffers from high sensitivity to noisy and missing input and adversarial attacks. To mitigate these deficiencies of deep RL solutions, we suggest involving spiking...

The coexistence of opposing dynamical states in a single system is an important and often baffling behavior observed in many fields of science and engineering. An example of such behavior is the coexistence of macroscopic order and microscopic disorder in hierarchical systems, like lasers and semiconductors, illustrating Haken’s principle of synerg...

The development of spiking neural network simulation software is a critical component enabling the modeling of neural systems and the development of biologically inspired algorithms. Existing software frameworks support a wide range of neural functionality, software abstraction levels, and hardware devices, yet are typically not suitable for rapid...

In this paper a random graph model [Formula presented] is introduced, which is a combination of fixed torus grid edges in (ℤ∕Nℤ)² and some additional random ones. The random edges are called long, and the probability of having a long edge between vertices u,v∈(ℤ∕Nℤ)² with graph distance d on the torus grid is pd = c∕Nd, where c is some constant. We...

Research in last few years on neurophysiology focused on several areas across the cortex during cognitive processing to determine the dominant direction of electrical activity. However, information about the frequency and direction of episodic synchronization related to higher cognitive functions remain unclear. Our aim was to determine whether neu...

Spiking neural networks are motivated from principles of neural systems and may possess unexplored advantages in the context of machine learning. A class of \textit{convolutional spiking neural networks} is introduced, trained to detect image features with an unsupervised, competitive learning mechanism. Image features can be shared within subpopul...

We present a system comprising a hybridization of self-organized map (SOM) properties with spiking neural networks (SNNs) that retain many of the features of SOMs. Networks are trained in an unsupervised manner to learn a self-organized lattice of filters via excitatory-inhibitory interactions among populations of neurons. We develop and test vario...

The development of spiking neural network simulation software is a critical component enabling the modeling of neural systems and the development of biologically inspired algorithms. Existing software frameworks support a wide range of neural functionality, software abstraction levels, and hardware devices, yet are typically not suitable for rapid...

Simulations of EEG data provide the understanding of how the limbic system exhibits normal and abnormal states of the electrical activity of the brain. While brain activity exhibits a type of homeostasis of excitatory and inhibitory mesoscopic neuron behavior, abnormal neural firings found in the seizure state exhibits brain instability due to runa...

There is an increasing interest in monitoring mental activities associated with different cognitive states using electroencephalography (EEG). In recent work, we introduced a comparative study of EEG recordings with participants in various cognitive modalities, including open eyes, open eyes with visual stimuli, closed eyes, and meditation. We show...

Average evoked potential data recorded as impulse responses of brains to electric shocks show Bessel-like functional distributions which we analyze in terms of couples of damped/amplified oscillators. This reproduces results obtained in terms of ordinary differential equations (Freeman K-sets) and offers the possibility of a direct connection with...

Electroencephalograph (EEG) data provide insight into the interconnections and relationships between various cognitive states and their corresponding brain dynamics, by demonstrating dynamic connections between brain regions at different frequency bands. While sensory input tends to stimulate neural activity in different frequency bands, peaceful s...

Measurements of local field potentials over the cortical surface and the scalp of animals and human subjects reveal intermittent bursts of beta and gamma oscillations. During the bursts, narrow-band metastable amplitude modulation (AM) patters emerge for a fraction of a second and ultimately dissolve to the broad-band random background activity. Th...