April 29, 2013 | A pair of researchers in Japan has used NVIDIA GPUs and the CUDA parallel programming model to create a 100,000-neuron simulation of the human cerebellum. And then they taught a robot to hit a baseball. The work was begun at the RIKEN Brain Science Institute, and was published online in Neural Networks last February.
One of the biggest challenges in modeling neural brain function is simulation speed. Using a CPU alone it took the researchers 98 seconds of compute time to figure out how to respond to a stimulus lasting just one second. Using GPUs resulted in a 100x speedup, giving the GPU-based system the speed needed to handle real world tasks. NVIDIA Blog