Machine learning’s rise has been facilitated by the use of graphic processing units (GPUs) instead of the more traditional central processing units (CPUs). GPUs are better than CPUs because they are designed to do their processing in parallel, says Werner Coetzee, Business Development Executive at Data Sciences Corporation.

“Although GPUs can process in parallel and thus move through multiple tasks more rapidly, they don’t offer the same level of precision as a CPU, but what the GPU does is it serves as a kind of turbo boost for the computer,” he says.