RESUMEN
The brain is dynamic, associative, and efficient. It reconfigures by associating the inputs with past experiences, with fused memory and processing. In contrast, AI models are static, unable to associate inputs with past experiences, and run on digital computers with physically separated memory and processing. We propose a hardware-software co-design, a semantic memory-based dynamic neural network using a memristor. The network associates incoming data with the past experience stored as semantic vectors. The network and the semantic memory are physically implemented on noise-robust ternary memristor-based computing-in-memory (CIM) and content-addressable memory (CAM) circuits, respectively. We validate our co-designs, using a 40-nm memristor macro, on ResNet and PointNet++ for classifying images and three-dimensional points from the MNIST and ModelNet datasets, which achieves not only accuracy on par with software but also a 48.1 and 15.9% reduction in computational budget. Moreover, it delivers a 77.6 and 93.3% reduction in energy consumption.
RESUMEN
We report the chemical vapor deposition (CVD) growth, characterization, and low-temperature magnetotransport of 1T phase multilayer single-crystalline VTe2 nanoplates. The transport studies reveal that no sign of intrinsic long-range ferromagnetism but localized magnetic moments exist in the individual multilayer metallic VTe2 nanoplates. The localized moments give rise to the Kondo effect, evidenced by logarithmical increment of resistivity with decreasing temperature and negative magnetoresistance (NMR) regardless of the direction of magnetic field at temperatures below the resistivity minimum. The low-temperature resistivity upturn is well described by the Hamann equation, and the NMR at different temperatures, a manifestation of the magnetization of the localized spins, is well fitted to a Brillouin function for S = 1/2. Density functional theory calculations reveal that the localized magnetic moments mainly come from the interstitial vanadium ions in the VTe2 nanoplates. Our results will shed light on the study of magnetic properties, strong correlation, and many-body physics in two-dimensional metallic transition metal dichalcogenides.