Deep learning has made substantial breakthroughs in many fields due to its powerful automatic representation capabilities. It has been proven that neural architecture design is crucial to the feature representation of data and the final performance. However, the design of the neural architecture heavily relies on the researchers’ prior knowledge and experience. And due to the limitations of humans’ inherent knowledge, it is difficult for people to jump out of their original thinking paradigm and design an optimal model. Therefore, an intuitive idea would be to reduce human intervention as much as possible and let the algorithm automatically design the neural architecture. Neural Architecture Search ( NAS ) is just such a revolutionary algorithm, and the related research work is complicated and rich. Therefore, a comprehensive and systematic survey on the NAS is essential. Previously related surveys have begun to classify existing work mainly based on the key components of NAS: search space, search strategy, and evaluation strategy. While this classification method is more intuitive, it is difficult for readers to grasp the challenges and the landmark work involved. Therefore, in this survey, we provide a new perspective: beginning with an overview of the characteristics of the earliest NAS algorithms, summarizing the problems in these early NAS algorithms, and then providing solutions for subsequent related research work. In addition, we conduct a detailed and comprehensive analysis, comparison, and summary of these works. Finally, we provide some possible future research directions.

Resistance training had positive effects on the executive cognitive ability and global cognitive function among the elderly; however, it had a weak-positive impact on memory. No significant improvement was found in attention. Triweekly resistance training has a better effect on general cognitive ability than biweekly. Further studies are needed focusing on the development and application of resistance training among the elderly.

No abstract

A systematic high-pressure study of the CdN x (x = 2, 3, 4, 5, and 6) system is performed by using the first-principles calculation method in combination with the particle swarm optimization algorithm. We proposed four stable high-pressure phases (P4mbm-CdN2, Cmmm-CdN4, I4̅2d-CdN4, and C2/c-CdN5) and one metastable high-pressure phase (C2/m-CdN6), for which the structural frames are composed of a diatomic quasi-molecule N2, standard armchair N-chain, S-type bent armchair N-chain, zigzag–antizigzag N-chain, and N14 network structure. Among them, the novel zigzag–antizigzag N-chain and N14 network structure are reported for the first time. More importantly, Cmmm-CdN4 and C2/m-CdN6 possess high stability under ambient conditions, which may be quenched to ambient conditions once they are synthesized at high-pressure conditions. The high decomposition energy barrier (1.14 eV) results in a high decomposition temperature (2500 K) of Cmmm-CdN4, while a low decomposition energy barrier (0.19 eV) results in a mild decomposition temperature (500 K) of C2/m-CdN6. The high energy density and outstanding explosive performance make Cmmm-CdN4, I4̅2d-CdN4, C2/c-CdN5, and C2/m-CdN6 potential high-energy materials. The electronic structure analyses show that these predicted high-pressure structures are all metallic phases, and the N–N and Cd–N bonds are the strong covalent and ionic bond interactions, respectively. The charge transfer from the Cd atom plays an important role in the stability of the proposed structures.