资源环境科技发展态势分析平台

High-energy electron scattering that can isolate pairs of nucleons in high-momentum configurations reveals a transition to spin-independent scalar forces at small separation distances, supporting the use of point-like nucleon models to describe dense nuclear systems.

The strong nuclear interaction between nucleons (protons and neutrons) is the effective force that holds the atomic nucleus together. This force stems from fundamental interactions between quarks and gluons (the constituents of nucleons) that are described by the equations of quantum chromodynamics. However, as these equations cannot be solved directly, nuclear interactions are described using simplified models, which are well constrained at typical inter-nucleon distances(1-5) but not at shorter distances. This limits our ability to describe high-density nuclear matter such as that in the cores of neutron stars(6). Here we use high-energy electron scattering measurements that isolate nucleon pairs in short-distance, high-momentum configurations(7-9), accessing a kinematical regime that has not been previously explored by experiments, corresponding to relative momenta between the pair above 400 megaelectronvolts per c (c, speed of light in vacuum). As the relative momentum between two nucleons increases and their separation thereby decreases, we observe a transition from a spin-dependent tensor force to a predominantly spin-independent scalar force. These results demonstrate the usefulness of using such measurements to study the nuclear interaction at short distances and also support the use of point-like nucleon models with two- and three-body effective interactions to describe nuclear systems up to densities several times higher than the central density of the nucleus.

Memristor-enabled neuromorphic computing systems provide a fast and energy-efficient approach to training neural networks(1-4). However, convolutional neural networks (CNNs)-one of the most important models for image recognition(5)-have not yet been fully hardware-implemented using memristor crossbars, which are cross-point arrays with a memristor device at each intersection. Moreover, achieving software-comparable results is highly challenging owing to the poor yield, large variation and other non-ideal characteristics of devices(6-9). Here we report the fabrication of high-yield, high-performance and uniform memristor crossbar arrays for the implementation of CNNs, which integrate eight 2,048-cell memristor arrays to improve parallel-computing efficiency. In addition, we propose an effective hybrid-training method to adapt to device imperfections and improve the overall system performance. We built a five-layer memristor-based CNN to perform MNIST10 image recognition, and achieved a high accuracy of more than 96 per cent. In addition to parallel convolutions using different kernels with shared inputs, replication of multiple identical kernels in memristor arrays was demonstrated for processing different inputs in parallel. The memristor-based CNN neuromorphic system has an energy efficiency more than two orders of magnitude greater than that of state-of-the-art graphics-processing units, and is shown to be scalable to larger networks, such as residual neural networks. Our results are expected to enable a viable memristor-based non-von Neumann hardware solution for deep neural networks and edge computing.