Submission Deadline: 01 June 2026 View: 411 Submit to Special Issue
Prof. Dr. Esteban Tlelo-Cuautle
Email: etlelo@inaoep.mx
Affiliation: Department of Electronics, Instituto Nacional de Astrofísica, Optica y Electrónica, Tonantinztla, Puebla 72840, Mexico
Research Interests: modeling and simulation, fractional circuits and systems, hardware security, machine learning, internet of things, optimization
Dr. Luis Gerardo de la Fraga
Email: fraga@cs.cinvestav.mx
Affiliation: Computer Science Department, CINVESTAV, Mexico City 07360, Mexico
Research Interests: graphing, hardware security, modeling and simulation, fractional circuits and systems, computer vision, metaheuristics, neural networks
Nowadays, dynamical systems face the challenges of being optimized and implemented on hardware. In addition, dynamical models intensify these challenges when they have fractional or variable orders, which require the development of improved computational models to more accurately approximate real-world behavior. Machine learning (ML) techniques, such as neural networks and physics-informed learning, provide efficient ways to model complex dynamics while reducing computational costs. Numerical simulation and appropriate algorithms for optimization can facilitate the emulation of dynamical systems based on memory devices, as well as the synthesis of fractional models with modern semiconductor devices. ML-driven approaches, including surrogate modeling and reinforcement learning, can automate parameter tuning and enhance system performance. For instance, chaotic and neuromorphic circuits and systems embed memory devices such as memristors, memcapacitors, memductors, and memtranstors. Data-driven modeling using ML can capture the nonlinear behavior of these devices, enabling more accurate and efficient hardware design. The design possibilities using these memory devices present challenges, both in their synthesis with modern semiconductor technology and in their application to artificial intelligence (AI) and machine learning in this new era.
This special issue is dedicated to presenting state-of-the-art results on the development of computational modeling and simulation of dynamical systems. A key focus is on machine learning-based techniques for system identification, prediction, and optimization of fractional-order dynamics. New algorithmic methods are required for the optimization and hardware design of fractional circuits and systems, which can embed memory devices to improve applications in hardware security, machine learning, internet of things, and neuromorphic systems. Deep learning and generative AI, for example, can accelerate the exploration of design spaces and improve simulation accuracy. Those designs can be performed in the analog, digital, and mixed-mode domains. ML-enhanced tools can bridge behavioral modeling with circuit-level implementation, reducing development time. Modern dynamical systems embed memory devices such as memristors, memcapacitors, memductors, and memtranstors. By integrating ML with traditional simulation, these systems can achieve adaptive learning and real-time optimization. These devices improve the way neuromorphic systems can be applied in security, artificial intelligence (AI) in the Internet of Things (IoT), Healthcare IoT, image processing, and so on. The development of algorithms for modeling, simulation, and optimization, generative AI modeling and simulation of chaotic and neuromorphic systems, play an important role in the modern design methods that range from behavioral descriptions down to circuit design and layout generation.
Suggested themes are listed as below:
· Machine learning-based methods
· Computational modeling
· Numerical simulation and optimization
· Neuromorphic circuits and systems
· Chaotic systems
· Fractional circuits and systems
· Machine learning for IoT, AIoT and HIoT
· Memristor, memcapacitor, memductor and memtranstor
· Conventional artificial and spiking neural networks
· Integrated circuits, embedded and FPGA devices
· Hardware security
Login
Register
Submit a Paper
Propose a Special lssue