A Vedic Mathematics-Based Approach to Enhancing Machine Learning Efficiency

Abstract

This article presents a structured framework for integrating Vedic Mathematics—a classical Indian system of mental calculation—into modern machine learning (ML) pipelines to improve computational efficiency, reduce energy consumption, and accelerate model training and inference. By mapping Vedic sutras to arithmetic-intensive operations in ML, such as matrix multiplication, convolution, and quantization-aware scaling, we demonstrate how hybrid Vedic–ML systems can achieve faster throughput, lower latency, and better scalability without compromising accuracy. The discussion covers algorithmic design, hardware implications, empirical evidence from recent studies, and practical guidelines for implementation.