Neuromorphic computing will be great… if hardware can handle the workload

Nanowerk  November 7, 2023 Ramp-reversal memory has recently been discovered in several insulator-to-metal transition materials where a non-volatile resistance change can be set by repeatedly driving the material partway through the transition. An international team of researchers (USA – Purdue University, University of Colorado, UCSD, France) has successfully developed a single-photon light source consisting of doped ytterbium ions (Yb3+) in an amorphous silica optical fiber at room temperature. They used optical microscopy to track the location and internal structure of accumulated memory as a thin film of VO2 was temperature cycled through multiple training subloops. The measurements revealed that the […]

Tiny memory cell withstands extreme temperatures

Nanowerk  October 16, 2023 Analog switching in ferroelectric devices promises neuromorphic computing with the highest energy efficiency if limited device scalability can be overcome. Researchers in Germany demonstrated the ferroelectric switching characteristics of sub-5 nm thin Al0.74Sc0.26N films grown on Pt/Ti/SiO2/Si and epitaxial Pt/GaN/sapphire templates by sputter-deposition. They focused on the following major achievements compared to previously available wurtzite-type ferroelectrics: 1) Record low switching voltages down to 1 V are achieved, which was in a range that could be supplied by standard on-chip voltage sources. 2) Compared to the previously investigated deposition of ultrathin Al1−xScxN films on epitaxial templates, a […]

Quantum material exhibits ‘non-local’ behavior that mimics brain function

Phys.org  August 8, 2023 A key aspect of how the brain learns and enables decision-making processes is through synaptic interactions. Electrical transmission and communication in a network of synapses are modulated by extracellular fields generated by ionic chemical gradients. Emulating such spatial interactions in synthetic networks can be of potential use for neuromorphic learning and the hardware implementation of artificial intelligence. A team of researchers in the US (Rutgers University, UC San Diego, Purdue University, Brookhaven National Laboratory, Argonne National Laboratory) has demonstrated that in a network of hydrogen-doped perovskite nickelate devices, electric bias across a single junction can tune […]

Research team takes neuromorphic computing a step forward

Phys.org  July 24, 2023 Magnons undergo nonlinear multimode scattering processes at large input powers. In experiments and simulations, an international team of researchers (Germany, France) showed that the interaction between magnon modes of a confined magnetic vortex could be harnessed for pattern recognition. They studied the magnetic response to signals comprising sine wave pulses with frequencies corresponding to radial mode excitations. The amplitudes of three-magnon scattering results in the excitation of different azimuthal modes depended strongly on the input sequences. They showed that recognition rates as high as 99.4% could be attained for four-symbol sequences using the scattered modes, with […]

Researchers develop fluidic memristor with diverse neuromorphic functions

Nanowerk  January 13, 2023 Mimicking the structure of human brain could lead to the development of next-generation neuromorphic devices. Focusing on different aspects of neuromorphic engineering, researchers in China created nanofluidic devices consisting of nanometer-thick two-dimensional slits filled with a salt solution, and a nanofluidic ionic memristor based on confined polyelectrolyte-ion interactions. They demonstrated neuromorphic functions with a polyelectrolyte-confined fluidic memristor (PFM), resulting in ion memory effects emulating various electric pulse patterns with ultralow energy consumption. PFM is versatile and easily interfaces with biological systems, paving a way to building neuromorphic devices with advanced functions by introducing rich chemical designs…read […]

Researcher is studying materials whose traits resemble those of the human brain

Phys.org  August 3, 2022 An international team of researchers (USA – Purdue University, New York University, UC San Diego, University of Chicago, Argonne National Laboratory, Northwestern, UC Santa Barbara, NIST, UC Davis, Brookhaven National Laboratory, France) summarizes and reflects on efforts to find “quantum materials” that can mimic brain function. CMOS has been engineered to keep different information states well-separated. It is not very well-designed for doing things where there is a lot of randomness and fluctuations. The human brain, on the other hand, can easily handle such tricky tasks while consuming dramatically less energy than modern computers. According to […]

Teaching robots to think like us

Science Daily  October 26, 2021 Researchers in Japan have taught a robot to navigate through a maze by electrically stimulating a culture of brain nerve cells connected to the machine. The neurons were grown from living cells and acted as the physical reservoir for the computer to construct coherent signals. The signals are regarded as homeostatic signals, telling the robot the internal environment was being maintained within a certain range and acting as a baseline as it moved freely through the maze. Throughout trials, the robot was continually fed the homeostatic signals interrupted by the disturbance signals until it had […]

New DNA-based chip can be programmed to solve complex math problems

Science Daily  September 14, 2021 The reactions in DNA-based combinational logic computing are mostly achieved through a manual process. For DNA-based Boolean logic, researchers in South Korea have fabricated a DNA-based microfluidic processing unit (MPU). They cast polydimethylsiloxane using double-sided molding techniques for alignment between the microfluidics and valve switch. For a uniform surface, molds fabricated using a three-dimensional printer were spin-coated by a polymer. For programming control, the valve switch arms were operated by servo motors. In the MPU controlled via a personal computer or smartphone application, the molecules with two input DNAs and a logic template DNA were […]

Towards next-gen computers: Mimicking brain functions with graphene-diamond junctions

Science Daily  August 4, 2021 Researchers in Japan designed graphene-diamond junctions that can mimic the characteristics of biological synapses and key memory functions, opening the doors for next-generation image sensing memory devices. They demonstrated optoelectronically controlled synaptic functions using junctions between vertically aligned graphene (VG) and diamond. The fabricated junctions mimic biological synaptic functions when stimulated with optical pulses and exhibit other basic brain functions such as the transition from short-term memory (STM) to long-term memory (LTM). The VG-diamond arrays underwent redox reactions induced by fluorescent light and blue LEDs under a bias voltage. The researchers attributed this to the […]