Apple paid $200 million for the startup, making neural network computing

Apple paid about $ 200 million in startup Xnor.ai, which develops enabling technologies for “boundary vysilani” (edge computing). This was announced by the publication GeekWire and confirmed to TechCrunch sources close to Apple.

development of the Xnor.ai is associated with an increase in the efficiency of algorithms for machine computation to such an extent that they could run on the weakest hardware up to chip, powered by photovoltaic cells. For example, decisions a startup can create cameras, solving the problem of recognition of objects independently, without relying on servers in the cloud and transfer vast amounts of data.

Xnor.ai during his work received investments in the amount of 14.7 million dollars, writes TechCrunch. So investors are well-earned, even if Apple paid amount is significantly less than the one reported by GeekWire.

For Apple technology boundary calculations are extremely important since they help to improve the efficiency of smart algorithms of camera and functions like speech recognition. In addition, they help to avoid sending information to the cloud, what is important from the point of view Apple proclaimed the priority of confidentiality and protection of user data.

Modern smartphones don’t need such tasks. But smart glasses with augmented reality functions (which are now probably develop in the Apple) will be able to become successful just by being light and graceful. Therefore, the account will be every milliamp-hour of miniature batteries, and neural network computing will need to make as energy efficient as possible.

Text: To.Hi-tech