Most existing person re-identification (ReID) methods have good feature representations to distinguish pedestrians with deep convolutional neural network (CNN) and metric learning methods. However, these works concentrate on the similarity between encoder output and ground-truth, ignoring the correlation between input and encoder output, which affects the performance of identifying different pedestrians. To address this limitation, We design a Deep InfoMax (DIM) network to maximize the mutual information (MI) between the input image and encoder output, which doesn't need any auxiliary labels. To evaluate the effectiveness of the DIM network, we propose end-to-end Global-DIM and Local-DIM models. Additionally, the DIM network provides a new solution for cross-dataset unsupervised ReID issue as it needs no extra labels. The experiments prove the superiority of MI theory on the ReID issue, which achieves the state-of-the-art results.

Cyber security has grown up to be a hot issue in recent years. How to identify potential malware becomes a challenging task. To tackle this challenge, we adopt deep learning approaches and perform flow detection on real data. However, real data often encounters an issue of imbalanced data distribution which will lead to a gradient dilution issue. When training a neural network, this problem will not only result in a bias toward the majority class but show the inability to learn from the minority classes.In this paper, we propose an end-to-end trainable Tree-Shaped Deep Neural Network (TSDNN) which classifies the data in a layer-wise manner. To better learn from the minority classes, we propose a Quantity Dependent Backpropagation (QDBP) algorithm which incorporates the knowledge of the disparity between classes. We evaluate our method on an imbalanced data set. Experimental result demonstrates that our approach outperforms the state-ofthe-art methods and justifies that the proposed method is able to overcome the difficulty of imbalanced learning. We also conduct a partial flow experiment which shows the feasibility of real-time detection and a zero-shot learning experiment which justifies the generalization capability of deep learning in cyber security.

Accurate real-time forecasts of inundation depth and extent during typhoon flooding are crucial to disaster emergency response. To manage disaster risk, the development of a flood inundation forecasting model has been recognized as essential. In this paper, a forecasting model by integrating a hydrodynamic model, k-means clustering algorithm and support vector machines (SVM) is proposed. The task of this study is divided into four parts. First, the SOBEK model is used in simulating inundation hydrodynamics. Second, the k-means clustering algorithm classifies flood inundation data and identifies the dominant clusters of flood gauging stations. Third, SVM yields water level forecasts with 1–3 h lead time. Finally, a spatial expansion module produces flood inundation maps, based on forecasted information from flood gauging stations and consideration of flood causative factors. To demonstrate the effectiveness of the proposed forecasting model, we present an application to the Yilan River basin, Taiwan. The forecasting results indicate that the simulated water level forecasts from the point forecasting module are in good agreement with the observed data, and the proposed model yields the accurate flood inundation maps for 1–3 h lead time. These results indicate that the proposed model accurately forecasts not only flood inundation depth but also inundation extent. This flood inundation forecasting model is expected to be useful in providing early flood warning information for disaster emergency response.

No abstract

To address semi-supervised learning from both labeled and unlabeled data, we present a novel meta-learning scheme. We particularly consider that labeled and unlabeled data share disjoint ground truth label sets, which can be seen tasks like in person re-identification or image retrieval. Our learning scheme exploits the idea of leveraging information from labeled to unlabeled data. Instead of fitting the associated class-wise similarity scores as most meta-learning algorithms do, we propose to derive semantics-oriented similarity representations from labeled data, and transfer such representation to unlabeled ones. Thus, our strategy can be viewed as a self-supervised learning scheme, which can be applied to fully supervised learning tasks for improved performance. Our experiments on various tasks and settings confirm the effectiveness of our proposed approach and its superiority over the state-of-the-art methods.