Short Communication: GPU-Aware Cost Models for ML Compilers
Abstract
Introducing a GPU instruction-level cost model improving compilation time and kernel fusion efficiency for TensorFlow XLA.
Cite this article
Orlov, S., Weiss, V., & Yilmaz, V. (2023). Short Communication: GPU-Aware Cost Models for ML Compilers. Research Explorations in Global Knowledge & Technology (REGKT), 2 (10). Retrieved from https://regkt.com/article.php?id=495&slug=short-communication-gpu-aware-cost-models-ml-compilers