AI Model Compression Techniques: Building Cheaper, Faster, and Greener AI

In this course, Tejas Chopra—an advocate for efficient, green computing—explores how to make AI/ML models more efficient, cost-effective, and environmentally friendly. Dive into practical techniques such as pruning, quantization, and knowledge distillation. Learn how to reduce model size and memory usage without significantly compromising accuracy. Use hands-on coding exercises in TensorFlow and PyTorch to implement these techniques and fine-tune models for optimal performance. Build your understanding of how to balance accuracy, efficiency, and sustainability, giving you the tools to build smarter, faster, and greener AI systems. Whether you’re building edge AI applications, deploying models at scale, or seeking to lower carbon footprints, this course equips you with actionable strategies to address real-world challenges in AI.

Learning objectives
Explain the motivations for model compression in AI/ML and its benefits in terms of memory usage, cost savings, sustainability, and real-world applications like edge deployment.
Understand and apply key model compression techniques such as pruning, quantization, and knowledge distillation, and evaluate their trade-offs.
Implement model compression techniques in TensorFlow and PyTorch through hands-on coding exercises, and measure their impact on model size, inference speed, and accuracy.
Evaluate the environmental and business impact of model compression by analyzing how compressed models reduce carbon footprints and operational costs.