GPGPU computing
There are several reasons that led to deep learning (DL) being developed and placed at the center of attention in the field of machine learning (ML) in the recent decades.
One reason, perhaps the main one, is surely represented by the progress in hardware, with the availability of new processors, such as Graphics Processing Units (GPUs), which have greatly reduced the time needed to train networks, reducing the time by 10 or even 20 times.
In fact, since the connections between the individual neurons have a numerically estimated weight, and since networks learn by calibrating the weights appropriately, increasing network complexity would cause high computing power, and high computing power can be handled by GPU.
The GPGPU history
GPGPU is an acronym that stands for General Purpose Computing on Graphics Processing Units. It recognizes the trend of employing GPU technology for applications other than graphics. Until 2006, the graphics API OpenGL and DirectX standards were the...
