Search

This study used the coordinated development of the regional economy and ecological environment in the Yellow River basin as the research object and selected 13 indicators representing the regional economic subsystem that reflect economic scale, economic structure, and economic efficiency in order to explore a path of high-quality economic development suitable for China’s national conditions and realize the modernization of China with common prosperity for all people. The coupling weights of each indicator were then calculated separately, along with the degree of orderliness between the two subsystems of the regional economy and ecological environment. Finally, the degree of synergy between the two subsystems was calculated. The subjective weights and objective weights of each indicator were first estimated using hierarchical analysis (AHP) and conflicting correlation among criteria (CRITIC). The findings reveal a strong growth tendency in the regional economic development of the Yellow River basin and a wave-like growth pattern in ecological environment indicators. Since it overtook the ecological economy in 2017, the orderliness of the regional economic subsystem has continued this growth trend, demonstrating that the regional economy is in front of the ecological environment and is in a dominant position. The Yellow River basin’s composite system of regional economic and natural environment exhibits a low-level synergy stage and an M-shaped development pattern of alternating increases and decreases. As a result, many solutions are suggested, including promoting high-quality growth with greater levels of opening, modifying the regional economic structure, and stepping up ecological environmental protection.

The ability to incrementally learn new classes is crucial to the development of real-world artificial intelligence systems. In this paper, we focus on a challenging but practical few-shot class-incremental learning (FSCIL) problem. FSCIL requires CNN models to incrementally learn new classes from very few labelled samples, without forgetting the previously learned ones. To address this problem, we represent the knowledge using a neural gas (NG) network, which can learn and preserve the topology of the feature manifold formed by different classes. On this basis, we propose the TOpology-Preserving knowledge InCrementer (TOPIC) framework. TOPIC mitigates the forgetting of the old classes by stabilizing NG's topology and improves the representation learning for few-shot new classes by growing and adapting NG to new training samples. Comprehensive experimental results demonstrate that our proposed method significantly outperforms other state-of-the-art class-incremental learning methods on CIFAR100, miniImageNet, and CUB200 datasets., Accepted by CVPR 2020 (oral)