Green AI for Electronic Design Automation

Talk title: Green AI for Electronic Design Automation: Sustainable Deep Neural Networks for Efficient Hardware and Systems Design

The rapid growth of deep neural networks has led to unprecedented advances in electronic systems, but at the cost of increasing energy consumption, carbon footprint, and hardware complexity. This training addresses these challenges by introducing Green AI principles applied to Electronic Design Automation (EDA), with a focus on sustainable, energy-efficient deep learning models and systems.

The training explores how deep neural networks can be designed, optimized, and deployed with sustainability as a first-class objective, spanning multiple abstraction levels from algorithmic design to hardware-aware optimization and system-level evaluation. Participants will learn how EDA methodologies and CAD tools can support energy-aware modeling, analysis, and verification of AI-enabled electronic systems.

Key topics include energy-efficient neural network architectures, model compression and approximation techniques, hardware software co-design, and sustainability-driven metrics for evaluating AI systems. The training also highlights emerging research directions where machine learning is used to enhance EDA workflows, enabling greener design choices throughout the electronic system lifecycle.

By bridging deep learning, sustainability, and electronic design automation, this training equips participants with the knowledge and tools needed to design next-generation AI systems that are both high-performing and environmentally responsible, in line with IEEE CEDA's mission to advance efficient and reliable electronic design.