Also known as: Graphical processing units Β· GPU
Graphics Processing Unit (GPU) is a specialized processor originally designed for rendering graphics and now widely used for parallel computation. In AI, GPUs accelerate matrix and tensor operations required for training and running machine learning models.
These are similar to central processing units, found on a typical home computer. Graphical processing units have been used since the 1970s in gaming applications and have been designed to accelerate computer graphics and image processing. In the past decade, graphical processing units have been increasingly applied in the training of large machine learning models after they were found to be effective in processing large amounts of data in parallel. For more details see PB 57.