+44 (0)24 7671 8970
More publications     •     Advertise with us     •     Contact us
News Article

CEA-Leti reports Neural Network breakthrough


A team comprising CEA-Leti, CEA-List and two CNRS laboratories has published a new paper in Nature Communications presenting the first complete memristor-based Bayesian neural network implementation for a real-world task—classifying types of arrhythmia recordings with precise aleatoric and epistemic uncertainty.

Considering medical-diagnosis and other safety-critical, sensory-processing applications that require accurate decisions based on a small amount of noisy input data, the study notes that while Bayesian neural networks excel at such tasks because they provide predictive uncertainty assessment, their probabilistic nature requires increased use of energy and computation. The increase is caused by the fact that implementing the networks in hardware requires a random number generator to store the probability distributions, i.e. synaptic weights.

“Our paper presents, for the first time, a complete hardware implementation of a Bayesian neural network utilizing the intrinsic variability of memristors to store these probability distributions,” said Elisa Vianello, CEA-Leti chief scientist and co-author of the paper, “Bringing Uncertainty Quantification to the Extreme-Edge with Memristor-Based Bayesian Neural Networks”. “We exploited the intrinsic variability of memristors to store these probability distributions, instead of using random number generators.”

A second major challenge was that performing inference in the team’s approach requires massive parallel multiply-and-accumulate (MAC) operations.

“These operations are power-hungry when carried out on CMOS-based ASICs and field-programmable gate arrays, due to the shuttling of data between processor and memory,” Vianello said. “In our solution, we use crossbars of memristors that naturally implement the multiplication between the input voltage and the probabilistic synaptic weight through Ohm’s law, and the accumulation through Kirchhoff’s current law, to significantly lower power consumption.”

Reconciling Memristors and Bayesian Neural Networks

Co-author Damien Querlioz, a scientist associated with the University of Paris-Saclay, the French National Center for Scientific Research (CNRS) and the Center of Nanosciences and Nanotechnologies, said the team also had to reconcile the nature of memristors, whose statistical effects adhere to the laws of device physics, with Bayesian neural networks, in which these effects can take arbitrary shapes.

“This work overcomes that challenge with a new training algorithm – variational inference augmented by a ‘technological loss’ – that accommodates device non-idealities during the learning phase,” he said. “Our approach enables the Bayesian neural network to be compatible with the imperfections of our memristors.”

Uncertainty quantification involves the network's ability to identify unknown situations out-of-distributions.

If a traditional neural network trained to recognize cats and dogs is presented with an image of a giraffe, it “confidently misclassifies” it as a cat or a dog, Vianello said. “In contrast, a Bayesian neural network would respond, ‘I am not entirely sure what this is because I have never seen it.’ While this example is lighthearted, in critical environments like medical diagnosis, where incorrect predictions can have severe consequences, this uncertainty-capturing ability becomes crucial.”

This capability arises from the fact that synaptic values in Bayesian neural networks are not precise values, as in traditional neural networks, but rather probability distributions. Consequently, the output is also a probability distribution, providing information about its “certainty”.

CEA-Leti reports Neural Network breakthrough
Advantest targets NAND Flash/NVM market
Mycronic receives order for SLX mask writer
AMD inaugurates largest global design centre in India
Renesas unveils own 32-bit RISC-V CPU Core
ASM announces $300M US expansion
Purdue honours Boilermaker and innovative visionary in semiconductors
Amkor to open US advanced packaging and test facility
Applied Materials and CEA-Leti unveil Joint Lab
Agile Analog partners with sureCore on CryoCMOS Innovate UK project
Worldwide semiconductor revenue to grow 17% in 2024
USPTO unveils Semiconductor Technology Pilot Program
Nordic Semiconductor celebrates 40th anniversary
ASML to appoint Christophe Fouquet as President and Chief Executive Officer
Commission launches Chips Joint Undertaking under the European Chips Act
Cadence aids Samsung Foundry’s 5G 'breakthrough success'
Large-scale commercial iSIM deployment
Resonac participates in US-based semiconductor consortium
DOE provides $42 million for power projects
SIPLACE TX micron from ASMPT combines SMT assembly and die processing
TRI strengthens presence in Malaysia
Particle accelerator on a microchip
Exosens signs US agreement with Scientific Imaging
Water, Energy, Sustainability: Grundfos's Impact on the Semiconductor Sector
STMicroelectronics and ERG sign renewables agreement
Resonac plans Process R&D Centre in Silicon Valley
EtherCAT Semiconductor Technical Working Group publishes further standards
Wisconsin Oven ships horizontal quench systems
'Redefining' 5G experience
ASMPT upbeat after productronica 2023
Agile Analog appoints former GUC Europe President as VP of Sales
Imec grants public access to virtual fab
Search the news archive

To close this popup you can press escape or click the close icon.
Register - Step 1

You may choose to subscribe to the Silicon Semiconductor Magazine, the Silicon Semiconductor Newsletter, or both. You may also request additional information if required, before submitting your application.

Please subscribe me to:


You chose the industry type of "Other"

Please enter the industry that you work in:
Please enter the industry that you work in: