Advances in AI-accelerated modelling of electromagnetic and electronic devices and systems

The rapid evolution of edge computing, 5G/6G networks, AI accelerators, and high-performance integrated systems demands ultra-efficient, miniaturized electronic devices capable of real-time intelligence under severe power, size, and thermal constraints. Traditional VLSI and high-frequency electronic design workflows rely on computationally intensive simulations of the underlying Maxwell, Poisson, and semiconductor transport equations, manual iterations, and rule-based optimization, creating bottlenecks in exploring complex design spaces for advanced SoCs, RFICs, and edge AI hardware. Artificial intelligence (AI) and machine learning (ML)—including surrogates, physics-informed neural networks (PINNs), generative models, and hybrid AI-physics approaches—are transforming the computational physics and applied electromagnetics that underpin electronic engineering by accelerating EDA flows, enabling data-driven inverse design, discovering novel circuit topologies, and embedding intelligence directly into hardware for adaptive, low-power performance in next-generation devices.This Research Topic aims to highlight groundbreaking progress in AI-accelerated modelling, design, and optimization of advanced electronic devices and systems, with emphasis on VLSI circuits, high-frequency integrated components, edge AI hardware, and hybrid computing architectures, grounded in the device physics, electromagnetic theory, and multiscale transport phenomena that govern their behaviour. It showcases innovative algorithms, PINNs, surrogate models, generative AI, and hybrid frameworks that address key challenges in physics-based simulation, computational efficiency, multi-objective optimization, and system-level integration. Wave propagation in mmWave/THz structures, carrier transport in advanced semiconductor nodes, and electromagnetic coupling in integrated systems sit at the heart of the contribution, bringing together the computational physics, applied electromagnetics, and device-physics communities with AI researchers and electronic engineers. We expect this to accelerate the development of cutting-edge electronics for edge AI, mmWave/THz applications, high-performance computing, and smart embedded systemThis Research Topic provides a forward-looking overview of AI-driven advancements that integrate fundamental electromagnetic principles and semiconductor device physics with cutting-edge electronic devices and systems, focusing on VLSI, high-frequency circuits, and intelligent hardware. It covers methodologies, innovations, and applications in advanced electronics. Original research articles, reviews, and perspectives are invited on (but not limited to) the following integrated themes:-AI/ML for electronic circuit design, optimization, and synthesis integrated with advanced VLSI front-ends, edge AI hardware, and hybrid analog-digital computing-Machine learning surrogates and accelerated simulations of full-wave electromagnetic and TCAD solvers for high-frequency electronic devices, VLSI/RF EDA workflows, and integrated circuit modeling-Physics-informed neural networks and hybrid AI-physics frameworks for VLSI circuits, high-frequency electronic problems, mmWave/THz devices, and low-power AI accelerators-AI-accelerated intelligent electronic devices, reconfigurable hardware, and adaptive high-frequency systems for edge computing and embedded intelligence-Data-driven inverse design and electromagnetic-aware optimization of integrated circuits, passives, RFICs, VLSI components, and high-frequency electronic systems-AI applications in advanced electronic systems, edge AI hardware, hybrid analog-RF/digital computing, electromagnetic-aware VLSI, and next-generation SoCs-Cutting-edge integrations of AI with the underlying physics of electronic devices and systems for mmWave/THz electronics, VLSI EDA automation, adaptive high-frequency hardware, and intelligent edge platforms Topic editors: Jun-Jiat Tiang, Wei Hong Lim, Hameedur Rahman, Ali Ebrahimi, Gianluigi Zito Source: https://www.frontiersin.org/research-topics/78966/advances-in-ai-accelerated-modelling-of-electromagnetic-and-electronic-devices-and-systems