MIT News June 8, 2020 To make memristors an international team of researchers (USA – MIT, Lawrence Berkeley National Laboratory, IBM, China, South Korea) first fabricated a negative electrode out of silicon, a positive electrode by depositing a slight amount of copper, followed by a layer of silver. They sandwiched the two electrodes around an amorphous silicon medium patterning a millimeter-square silicon chip with tens of thousands of memristors. When they ran the chip through several visual tasks, the chip was able to “remember” stored images and reproduce them many times over, in versions that were crisper, and cleaner compared […]
Phys.org March 24, 2020 Memristive systems offer promising solutions for implementing novel in-memory computing architectures for machine learning and data analysis problems. An international team of researchers (Germany, Switzerland) argue that they are also ideal building blocks for integration in neuromorphic electronic circuits suitable for ultra-low power brain-inspired sensory processing systems. They present a recipe for creating such systems based on design strategies and computing principles inspired by those used in mammalian brains, enumerate the specifications and properties of memristive devices required to support always-on learning in neuromorphic computing systems and to minimize their power consumption. They discuss in what […]
University of Michigan, July 30, 2018 Memristors enable memory and processing in the same device. However, memristors can have resistances that are on a continuum. Researchers at the University of Michigan got around the problem by digitizing the current outputs and mapped large mathematical problems into smaller blocks within the array, called “memory-processing units,” improving the efficiency and flexibility of the system. This is particularly useful for implementing machine learning and artificial intelligence algorithms, weather prediction and other matrix-based operations… read more.