New chip reduces neural networks’ power consumption by up to 95 percent
New chip reduces neural networks’ power consumption by up to 95 percent - short science news

Neural networks are at the basis of many artificial-intelligence applications. However, this comes at a cost, since neural networks require a high amount of energy which makes them unsuitable for small, portable devices. Now, MIT researchers have developed a special-purpose chip that increases the speed of neural-network computations by three to seven times over its predecessors, while reducing power consumption 94 to 95 percent. This opens the possibility of running neural networks on smartphones and other small devices.

Read the full story: MIT