Neuromorphic computing: Self-correcting memristor array; tellurium nanowires; anomalous Hall torque.
Researchers at Korea Advanced Institute of Science and Technology (KAIST), Seoul National University, Sungkyunkwan University, Electronics and Telecommunications Research Institute (ETRI), and Yonsei University developed a memristor-based neuromorphic chip that can learn and correct errors, enabling it to adapt to immediate environmental changes.
The system consists of a selector-less 32 × 32 memristor crossbar array, peripheral circuitry, and digital controller that can run AI algorithms in the analog domain. The team says the system can learn to automatically separate a moving object from the background when processing a video stream and become better at the task over time. It achieved accuracy comparable to ideal computer simulations in real-time image processing.
According to the researchers, the device is ready for use in applications such as smart security cameras and medical devices that can analyze health data in real time. [1]
Researchers from Tohoku University, Suzhou University of Science and Technology, Purdue University, Beijing University of Posts and Telecommunications, University of Tennessee, Shanghai University, and Nanjing University demonstrated room-temperature ferroelectric and resistive switching behaviors in single-element tellurium (Te) nanowires.
They used these effects to build a self-gated ferroelectric field-effect transistor (SF-FET) that integrates both ferroelectric and semiconducting properties in a single device. The SF-FET had high data retention, fast switching speeds of less than 20 nanoseconds, and a storage density exceeding 1.9 terabytes per square centimeter.
“Our breakthrough opens up new opportunities for next-generation memory devices, where Te nanowires’ high mobility and unique electronic properties could help simplify device architectures,” said Yaping Qi, an assistant professor at Tohoku University’s Advanced Institute for Materials Research, in a release. “Our SF-FET device could also play a crucial role in future artificial intelligence systems, enabling neuromorphic computing that mimics human brain function. Additionally, the findings can help lead to lower power consumption in electronic devices, addressing the need for sustainable technology.” [2]
Researchers from the University of Utah and the University of California Irvine discovered a new type of spin-orbit torque that provides a new way to manipulate spin and magnetization through electrical currents.
Called anomalous Hall torque, it is related to the anomalous Hall effect. Instead of describing electron scattering, however, anomalous Hall torque describes how when an external electrical current is applied to a material, a spin current flows 90 degrees to the flow of electrical current with the spin-orientation along the direction of the magnetization. It complements spin Hall torque and planar Hall torque to form a triad of torques that should be present in all conductive spintronic materials.
“This is brand new physics, which on its own is interesting, but there’s also a lot of potential new applications that go along with it,” said Eric Montoya, assistant professor of physics and astronomy at the University of Utah, in a statement. “These self-generated spin-torques are uniquely qualified for new types of computing like neuromorphic computing, an emerging system that mimics human brain networks.”
“We utilized anomalous Hall torque to create a nanoscale device known as a spin-torque oscillator. This device can mimic the functionality of a neuron, but is significantly smaller and operates at higher speeds,” added Ilya Krivorotov, physicist at UCI, in a statement. “Our next step is to interconnect these devices into a larger network, enabling us to explore their potential for performing neuromorphic tasks, such as image recognition.” [3]
[1] Jeong, H., Han, S., Park, SO. et al. Self-supervised video processing with self-calibration on an analogue computing platform based on a selector-less memristor array. Nat Electron (2025). https://doi.org/10.1038/s41928-024-01318-6
[2] Zhang, J., Zhang, J., Qi, Y. et al. Room-temperature ferroelectric, piezoelectric and resistive switching behaviors of single-element Te nanowires. Nat Commun 15, 7648 (2024). https://doi.org/10.1038/s41467-024-52062-6
[3] Montoya, E.A., Pei, X. & Krivorotov, I.N. Anomalous Hall spin current drives self-generated spin–orbit torque in a ferromagnet. Nat. Nanotechnol. (2025). https://doi.org/10.1038/s41565-024-01819-7
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
Leave a Reply