EPJ ST Special Issue: Discrete neural networks: Firing patterns and synchronization strategies-basics for new AI technologies

Details

Published on 01 September 2023

The progress made in artificial intelligence has greatly influenced the everyday lives of humans. One of the key drivers behind AI development is the study of neural networks. Neural networks are capable of interpreting numerical patterns in the form of vectors and their primary function is to classify and categorize data based on similarities. Modern technologies such as self-driving cars, facial recognition, and language translation heavily rely on neural networks. These networks enable computers to make intelligent decisions with minimal human intervention because they can learn and model complex and nonlinear relationships between input and output data.

Analyzing neural networks and their dynamics involves studying equilibrium states, periodic solutions, stability, chaos, and optimization. Gaining insight into firing patterns is a crucial factor in understanding the dynamic features of neural networks. Firing patterns reflect the salient features of the stimuli that generate them. Therefore, investigating firing patterns is essential for uncovering hidden dynamics within network models. Additionally, the influence of time delays in neural networks, particularly in the context of mimicking the functioning of the brain, has attracted attention. Mathematical models illustrate numerous biological phenomena related to neuronal activities, including multi-stability, synchronization, spiking, and bursting patterns. Therefore, studying the dynamics of neural networks plays a significant role in understanding biological phenomena and designing practical procedures for information processing and signal propagation.

Synchronization of neural networks is a crucial phenomenon that provides insight into controlling the rhythmic activity within neuronal systems. Synchronization has been a subject of active research for a considerable period, achieved by implementing controllers that force a chaotic system (slave system) to track a master system. In practical applications, discrete iterative algorithms are commonly used as controllers, making it beneficial to consider the discrete time version of the model for analysis. Hence, this special issue is devoted to current state of art regarding “Discrete neural networks: Firing patterns and synchronization strategies”.

Topics covered include, but are not limited to:

• Discrete neural network models with and without delay.
• Coupled discrete neural networks with discrete memristor.
• Neural networks, fuzzy logic, and evolutionary based interpretable control systems
• Applications of neural computing in image/signal processing, business intelligence, games, healthcare, bioinformatics, security, robotics etc.
• Artificial intelligence, machine learning and big data.
• Mathematical analysis and computational analysis.
• Decision making, pattern recognition, optimization, forecasting, data analysis
• Experimental realization of discrete network models
• Stability, chaos, multi-stability and coexisting attractors in discrete models
• Stochastic discrete time neural networks
• Partial synchronization phenomena (for brain, Earth etc.)
• Synchronization in adaptive networks

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere. An extended description of the critical aspects/open problems of the methods presented will be a stringent criterion of pre-selection of papers to be sent to referees. Articles may be one of four types: (i) mini reviews (10-15 pages), (ii) tutorial reviews (15+ pages), (iii) original paper v1 (5-10 pages), or (iv) original paper v2 (3-5 pages). More detailed descriptions of each paper type can be found in the Submission Guidelines. Manuscripts should be prepared using the latex template (preferably single column layout), which can be downloaded here.

Articles should be submitted to the Editorial Office of EPJ ST via the submission system, and should be clearly identified as intended for the topical issue “Discrete neural networks: Firing patterns and synchronization strategies”.

Guest Editor:

Santo Banerjee, Department of Mathematical Sciences, Giuseppe Luigi Lagrange, Politecnico di Torino, Corso Duca degli Abruzzi, 24, Turin, Italy E-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.

Open Access: EPJST is a hybrid journal offering Open Access publication via the Open Choice programme and a growing number of Transformative Agreements which enable authors to publish OA at no direct cost (all costs are paid centrally).