Machine Learning-Driven Paradigms in VLSI Design: Exploring the Synergy with Pruning for Optimal Circuit Efficiency

This survey paper explores the intersection of machine learning and VLSI design, shedding light on the evolving landscape of innovative methodologies. The initial focus is on papers discussing the pivotal role of Machine Learning(ML) in shaping VLSI design paradigms. Subsequently, the paper delves into the realm of approximate computing, identifying pruning as a standout concept for its efficiency. Building on this, the paper advocates for a synergistic approach, proposing the integration of machine learning and pruning for optimal circuit design. Two illustrative papers are discussed to underscore the practical implications of this combined methodology, providing a holistic perspective on the advancements in VLSI design.