PAPER TITLE: A COMPREHENSIVE REVIEW OF ENERGY-EFFICIENT GENERATIVE AI: CHALLENGES, STRATEGIES, AND FUTURE DIRECTIONS
DOI-DS No: 03.2025-94499139
DOI Link: https://doi-ds.org/doilink/03.2025-94499139/waims.1412832
Author/s: Mr. Nitin Dhingra, Research Scholar University of Technology, Jaipur
Abstract: Generative AI models, including Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), and Transformer-based architectures, require extensive computational resources, leading to high energy consumption. As AI adoption grows, optimizing energy efficiency becomes crucial to reducing environmental impact and operational costs. This paper explores the key challenges in energy-efficient generative models and highlights recent innovations in model compression, hardware acceleration, and algorithmic optimizations. We also discuss trade-offs between energy efficiency and model performance, proposing sustainable strategies for future development.
Keywords: Generative AI, Energy Efficiency, Model Compression, Hardware Acceleration, Sustainable AI
| Full Paper Download |