Integrating Memristors For Neuromorphic Computing


Much of the current research on neuromorphic computing focuses on the use of non-volatile memory arrays as a compute-in-memory component for artificial neural networks (ANNs). By using Ohm’s Law to apply stored weights to incoming signals, and Kirchoff’s Laws to sum up the results, memristor arrays can accelerate the many multiply-accumulate steps in ANN algorithms. ANNs are being dep... » read more

What’s Next In Neuromorphic Computing


To integrate devices into functioning systems, it's necessary to consider what those systems are actually supposed to do. Regardless of the application, [getkc id="305" kc_name="machine learning"] tasks involve a training phase and an inference phase. In the training phase, the system is presented with a large dataset and learns how to "correctly" analyze it. In supervised learning, the data... » read more

Power/Performance Bits: Jan. 23


Atomristors for thin memory Engineers at The University of Texas at Austin and Peking University developed a thin memory storage device with dense memory capacity. Dubbed "atomristors," the device enables 3-D integration of nanoscale memory with nanoscale transistors on the same chip. "For a long time, the consensus was that it wasn't possible to make memory devices from materials that were... » read more

System Bits: Jan. 16


Nitrogen-atom-sized sensors A new quantum sensor developed by Fraunhofer researchers will be able to measure the tiny magnetic fields of the next generation of hard discs, leveraging the new opportunities that quantum technology promises. [caption id="attachment_430671" align="aligncenter" width="300"] The special ellipsoid form of the plasma reactor developed at Fraunhofer IAF allows for l... » read more

Toward Neuromorphic Designs


Part one of this series considered the mechanisms of learning and memory in biological brains. Each neuron has many fibers, which connect to adjacent neurons at synapses. The concentration of ions such as potassium and calcium inside the cell is different from the concentration outside. The cellular membrane thus serves as a capacitor. When a stimulus is received, the neuron releases neur... » read more

Power/Performance Bits: Nov. 8


Scrap metal batteries A research team at Vanderbilt University used scraps of steel and brass - two of the most commonly discarded materials - to create a steel-brass battery that can store energy at levels comparable to lead-acid batteries while charging and discharging at rates comparable to ultra-fast charging supercapacitors. The researchers found that when scraps of steel and brass a... » read more

Power/Performance Bits: Oct. 18


Speeding up memory with T-rays Scientists at the Moscow Institute of Physics and Technology (MIPT), the University of Regensburg in Germany, Radboud University Nijmegen in the Netherlands, and Moscow Technological University proposed a way to improve the performance of memory through using T-waves, or terahertz radiation, as a means of resetting memory cells. This process is several thousand... » read more

System Bits: May 3


Neural network synapses In a development that could potentially be used as a basis for the hardware implementation of artificial neural networks, Moscow Institute of Physics and Technology (MIPT) researchers have created prototypes of electronic synapses based on ultra-thin films of hafnium oxide (HfO2). The team made the HfO2-based memristors measuring just 40x40 nm2, which exhibit propert... » read more

Power/Performance Bits: Feb. 9


Molybdenum disulfide memristors Researchers at Michigan Technological University constructed an ideal memristor based on molybdenum disulfide nanosheets. "Different from an electrical resistor that has a fixed resistance, a memristor possesses a voltage-dependent resistance," said Yun Hang Hu, professor of materials science and engineering at MTU, adding that a material's electric propert... » read more

Power/Performance Bits: Oct. 20


Memristors come in threes The race is on to produce a commercial memristor, and a duo from ETH Zurich may be providing a bit more push. "Basically, memristors require less energy since they work at lower voltages," explained Jennifer Rupp, professor in the Department of Materials at ETH Zurich. "They can be made much smaller than today's memory modules, and therefore offer much greater de... » read more

← Older posts