Some of the best circuits to drive AI in the future may be analog, not digital, and research teams around the world are increasingly developing new devices to support such analog AI.

The most basic computation in the deep neural networks driving the current explosion in AI is the multiply-accumulate (MAC) operation. Deep neural networks are composed of layers of artificial neurons, and in MAC operations, the output of each one of these layers is multiplied by the values of the strengths or “weights” of their connections to the next layer, which then sums up these contributions.

Modern computers have digital components devoted to MAC operations, but analog circuits theoretically can perform these computations for orders of magnitude less energy. This strategy—known as analog AI, compute-in-memory or processing-in-memory—often performs these multiply-accumulate operations using non-volatile memory devices such as flash, magnetoresistive RAM (MRAM), resistive RAM (RRAM), phase-change memory (PCM) and even more esoteric technologies.

Machinia exists in the future when the world of robots is the norm. And, as has been highlighted in other posts, the world of the “extraordinary tomorrow” begins with the “unexceptional today.” #clean, #university

SAIT has started using autonomous robots to clean the floors at its main campus.

The robots are from Brain Corp, an artificial intelligence (AI) company, and contracted to SAIT for use through Calgary-based Bee Clean, Canada’s largest janitorial services provider.

In a Tuesday news release, Bee Clean said the robots will help maintain a “safe and clean” environment for students by scrubbing 58,000 square feet daily on the main campus.

Revolutionary new electronic components can be adapted to perform very different tasks – a technology perfectly suited for artificial intelligence.

Normally, computer chips consist of electronic components that always do the same thing. In the future, however, more flexibility will be possible: New types of adaptive transistors can be dynamically switched during run-time to perform different logical tasks. This fundamentally changes the possibilities of chip design and opens up completely new opportunities in the field of artificial intelligence, neural networks or even logic that works with more values than just 0 and 1.

In order to achieve this, scientists at TU Wien (Vienna) did not rely on the usual silicon technology, but on germanium. This was a success: The most flexible transistor in the world has now been produced using germanium. It has been presented in the journal ACS Nano. The special properties of germanium and the use of dedicated program gate electrodes made it possible to create a prototype for a new component that may usher in a new era of chip technology.