Climate change and rapid urbanization have led to the Urban Heat Island (UHI) effect, resulting in higher temperatures in metropolitan areas and negatively impacting urban communities. Accurate UHI forecasting is crucial for identifying high-risk periods and locations, especially in cities with vulnerable populations. Current methods are limited by data granularity and inadequate modeling of regional thermodynamics, which affects both accuracy and spatio-temporal granularity. In this paper, we propose DeepUHI, a data-driven context-aware framework for modeling local thermodynamics based on the heat equation, alongside the SeoulTemp dataset, the first multi-modal dataset for UHI effect predictions at the street level. Our framework utilizes a heat decomposition method to represent urban thermodynamics through thermodynamic cycles and thermal flows, effectively integrating urban environmental data. Extensive experiments show that our framework improves accuracy in UHI effect prediction and warning tasks, outperforming leading models. We have integrated DeepUHI into our SeoUHI platform to provide hourly street-level UHI forecasting for Seoul. The code, platform, and dataset are accessible at https://github.com/CityMind-Lab/DeepUHI.

In long-term time series forecasting (LTSF) tasks, an increasing number of works have acknowledged that discrete time series originate from continuous dynamic systems and have attempted to model their underlying dynamics. Recognizing the chaotic nature of real-world data, our model, Attraos, incorporates chaos theory into LTSF, perceiving real-world time series as low-dimensional observations from unknown high-dimensional chaotic dynamical systems. Under the concept of attractor invariance, Attraos utilizes non-parametric Phase Space Reconstruction embedding along with a novel multi-resolution dynamic memory unit to memorize historical dynamical structures, and evolves by a frequency-enhanced local evolution strategy. Detailed theoretical analysis and abundant empirical evidence consistently show that Attraos outperforms various LTSF methods on mainstream LTSF datasets and chaotic datasets with only one-twelfth of the parameters compared to PatchTST.

Complex-valued convolutional neural network (CV-CNN) has been presented in recent years. In this letter, CV full convolutional neural network (CV-FCNN) is proposed for synthetic aperture radar (SAR) target classification, which contains only convolution layers in the hidden layer. The purpose of replacing both the pooling and fully connected layers in CV-CNN with the convolution layers is to avoid complex pooling operation and prevent overfitting, respectively. Considering the label of target is always real-valued, the magnitude of the complex vector obtained from the last convolution layer is calculated before softmax classification in the output layer. Moreover, the back-propagation formula for each layer of CV-FCNN is presented in detail. Furthermore, the complex$1\times 1$convolution layer is added into CV-FCNN to learn the cross-channel information of feature maps. The experimental results show that the average accuracy can be improved using CV-FCNN, and it is further improved using CV-FCNN with the$1\times 1$convolution layer.