TDK, CEA team for AI spin memristor to slash power
Cette publication existe aussi en Français
TDK and CEA in France have developed a spin memristor for neuromorphic AI devices.
TDK worked with CEA and Tohoku University in Japan on the spin memristor for AI with a power consumption 1% of previous devices.
The spin memristor uses the charge and spin of electronics in an MRAM element and built prototype learning neuromorphic AI devices with the Centre for Innovative Integrated Electronic Systems (CIES) at Tohoku University.
In recent years, digital transformation (DX) has advanced due to the development of AI; it is predicted that energy consumption utilizing big data and AI will boom and make certain issues more apparent — such as the complexity around the computational processing of vast amounts of data, and the increased power consumption associated with the development of AI. TDK strives to contribute to solving these social and environmental issues.
First complete memristor neural network for medical AI
TDK set itself a goal to develop a device that electrically simulates the synapses of the human brain: the memristor. While conventional memory elements store data in the form of either 0 or 1, a spin memristor can store data in analog form, just as the brain does.
This makes it possible to perform complex computations with ultra-low power consumption. Although memristors for neuromorphic devices already exist, they face issues such as changes in resistance over time, difficulties in controlling the precise writing of data, and the need for control to ensure that data is retained. TDK’s spin-memristor solves these issues and is expected to offer immunity to environmental influences and long-term data storage while reducing power consumption by cutting leakage current in existing devices.
THe prototype AI circuit has 3 memristor elements × 2 sets × 4 chips and confirmed its successful operation through a sound separation demonstration, showing that the spin-memristors can serve as basic elements in AI circuits.
In the demonstration, even when three types of sound (music, speech, and noise were mixed with arbitrary ratios, the circuit was able to learn and separate the three types of sound in real-time. In general machine learning, AI operations are performed based on data that the AI model has previously been trained on, but TDK’s device is uniquely capable of learning in a changing environment in real-time.
The manufacturing of these products requires the integration of semiconductor and spintronic manufacturing processes. This integration has been achieved in the manufacturing of MRAM, a product similar to memristors, and TDK has decided to pursue integrated technological development on a joint basis with Tohoku University.
“The synergy between TDK and CEA is remarkable, as our complementary expertise fosters a highly creative and constructive collaboration. This research partnership is breaking new ground to develop more sustainable, reliable, highly efficient solutions that will meet the growing demands of modern AI applications,” said Marc Duranton, Senior Fellow of CEA.
“AI semiconductors are extremely important for the information-oriented society of the future, but the societal issues are improving AI processing power and reducing power consumption. In light of this social demand, TDK’s AI semiconductor development program, which fuses memristor and spintronics technology, is extremely important. We will do our best to contribute to this project with the academic knowledge held by Tohoku University and the manufacturing technology of the 12-inch prototype line,” said Tetsuo Endoh, the director of CIES Tohoku University.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
