GTC Silicon Valley-2019 ID:S9250:Implementing Machine Learning on GPUs
Many machine learning algorithms can benefit from the efficient fine-grained parallelism and high throughput of GPUs. Importantly, GPUs make it possible to complete training and inference much faster than possible on ordinary CPUs. In this talk, we will share some recent developments in implementing machine learning on GPUs and show how to leverage H2O.ai's H2O GPU edition to capture benefits from GPU acceleration. We'll also compare performance gains from GPUs with CPU performance.