AI‐Driven VLSI

With advancing technology, very-large-scale integration (VLSI) is leading the way with the integration of artificial intelligence (AI), machine learning (ML), and quantum computing into day-to-day applications. The integration of VLSI and AI has enabled revolutionary improvement in semiconductor technology, providing intelligent, high-speed, and low-power computing systems. Traditional VLSI design processes are typically constrained by manual optimization methodologies, larger design complexity, and vast resource utilization. The application of AI, in the guise of ML and deep learning (DL), has introduced automation to essential phases of VLSI design like circuit layout, optimization, testing, and verification. AI has been a need to propel electronic design automation (EDA), advancement, productivity, and precision in many aspects of the design process, with the increased complexity of new chip designs. AI-based techniques enhance power, performance, and area (PPA) trade-offs, reduce design iterations, and improve fault detection mechanisms, leading to higher yield and reduced time-to-market. Furthermore, AI predictive analytics in chip production optimizes defect detection and process control for better chip reliability and efficiency. AI models analyze vast amounts of historical design data and prior layouts to predict optimal floor plans and placements, minimizing congestion, reducing interconnect delays, and lowering power consumption. This chapter focuses on AI utilization in VLSI design, its application areas in EDA, and challenges in AI-based VLSI systems. On the basis of the review of recent AI techniques, upcoming trends, and illustrations, the book presents an exhaustive account of the combination of VLSI and AI and establishes a base for future innovation and research in smart semiconductor design.