Quantum machine learning is a promising application of quantum computing for data classification. However, most of the previous research focused on binary classification, and there are few studies on multi-classification. The major challenge comes from the limitations of near-term quantum devices on the number of qubits and the size of quantum circuits. In this paper, we propose a hybrid quantum neural network to implement multi-classification of a real-world dataset. We use an average pooling downsampling strategy to reduce the dimensionality of samples, and we design a ladder-like parameterized quantum circuit to disentangle the input states. Besides this, we adopt an all-qubit multi-observable measurement strategy to capture sufficient hidden information from the quantum system. The experimental results show that our algorithm outperforms the classical neural network and performs especially well on different multi-class datasets, which provides some enlightenment for the application of quantum computing to real-world data on near-term quantum processors.
The Yangtze River Basin has periodically been subject to torrential rains and floods. It is of great significance to characterize the extreme precipitation patterns to learn about frequent flood characteristics in the Yangtze River Basin. Commonly, spatiotemporal characteristics of extreme precipitation was studied by regional frequency analysis method with site data. Spatial sparse site data may cause imprecise divisions of homogeneous regions. In this paper, the spatiotemporal characteristics of extreme precipitation was studied by regional frequency analysis with corrected satellite-based grid precipitation data (Global Land Data Assimilation System, GLDAS) rather than site data. The results show that: (1) The corrected GLDAS daily precipitation data had greatly improved its ability to capture extreme precipitation events in Yangtze River Basin, as the data average accuracy increased from 0.215 before correction to 0.849 after correction. It is feasible to use satellite-based grid precipitation data to replace the site data for the regional frequency analysis of extreme precipitation. (2) The Yangtze River Basin was categorized into seven homogeneous regions for the annual maximum 1-day (RX1DAY) index with an automatic subjective adjustment method. (3) The regional growth curves and quantiles of the Yangtze River Basin were drawn for the return period for 2-100 years. (4) Spatial patterns of extreme daily precipitation series with a return period of 100 years indicated that the precipitation amount increases gradually from the upper to the lower Yangtze River Basin, from the "arid zone" to the "wet zone" and then to the "special wet zone", and the 100-year return level of RX1DAY varied from 30.3 to 301.8 mm. There were three main precipitation centres, the Sichuan Basin, Dongting Lake Basin, and a great triangle area covering the Poyang Lake Basin and the south foot of Dabie Mountain.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.