Improving Time Efficiency of Machine Learning Algorithms Through GPU Parallelization

This paper discusses the application of parallel computing technologies in artificial intelligence and machine learning processes. The study focuses on heterogeneous computing systems based on CPUs and GPUs, as well as the use of CUDA technology for parallel data processing. Experimental results show that GPU-based parallelization significantly improves computational speed and reduces execution time compared to traditional CPU-based processing. The research confirms the effectiveness of GPUs in accelerating machine learning algorithms and other computationally intensive tasks.