AI professionals have been pursuing a type of AI that could definitely bring down the energy needed to do average AI things like perceiving words and pictures. This analog type of machine learning does one of the key mathematical tasks of neural networks utilizing physics of a circuit rather than digital rationale. However, one of the principal things restricting this methodology is that deep learning's training algorithm, back propagation, must be finished by GPUs or other separate digital frameworks. A research collaboration between neuromorphic chip startup Rain Neuromorphics and Canadian research institute Mila has demonstrated that training neural networks utilizing totally analog hardware is conceivable, making the chance of end-to-end analog neural networks. This has significant impacts for neuromorphic processing and AI hardware overall – it guarantees completely analog AI chips that can be utilized for training and inference, making noteworthy savings on compute, power, latency and size.

This analog type of machine learning does one of the key mathematical … reconsider the hardware computational substrate for artificial intelligence, …

Nvidia stock is soaring, up more than 14% in after-hours trading, beating Wall Street's expectations for its first quarter. The graphics processing unit (GPU) company announced its first quarter earnings today with revenue of $1.94 billion, a 48% gain year over year. While gaming drives most of the company's sales, its largest area of growth over the past year was selling GPUs that power the artificial intelligence and graphics processing in datacenters. That sector of business saw 186% growth year over year, just a dip below last quarter's 205% growth. Nvidia has built itself into a leader in this space, selling enterprise GPUs to Google, Microsoft, Amazon, IBM, Tencent, and Baidu.