时间：2021.6.11 下午 2.30

地点：理科楼B311

标题：《Lightweight Feedback Convolution Neural Network for Remote Sensing Images Super-Resolution》

汇报人：吴一鸣

摘要：

There are lots of image data in the eld of remote sensing, most of which have low-resolution due to the limited image sensor. The super-resolution method can effectively restore the low-resolution image to the high-resolution image. However, the existing super-resolution method has both heavy computing burden and number of parameters. For saving costs, we propose the feedback ghost residual dense network (FGRDN), which considers the feedback mechanism as the framework to attain lower features through high-level rening. Further, for feature extraction, we replace the convolution of the residual dense blocks (RDBs) with ghost modules (GMs), which can remove the redundant channels and avoid the increase of parameters along with the network depth. Finally, the spatial and channel attention module (SCM) is employed in the end of the RDB to learn more useful information from features. Compared to other SOTA lightweight algorithms, our proposed algorithm can reach convergences more rapidly with fewer parameters, and the performance of the network can be markedly enhanced on the image texture and object contour reconstruction with better peak signal-to-noise ratio (PSNR) and structural similarity (SSIM).

录取期刊：ACCESS

标题：《Multiple Strategies Differential Privacy on Sparse Tensor Factorization for Network Traffic Analysis in 5G》

汇报人：韩惠

摘要：

Due to high capacity and fast transmission speed, 5G plays a key role in modern electronic infrastructure. Meanwhile, Sparse Tensor Factorization (STF) is a useful tool for dimension

reduction to analyze High-Order, High-Dimension, and Sparse Tensor (HOHDST) data which is transmitted on 5G Internetof- things (IoT). Hence, HOHDST data relies on STF to obtain complete data and discover rules for real-time and accurate analysis. From another view of computation and data security, the current STF solution seeks to improve the computational efficiency but neglects privacy security of the IoT data, e.g., data analysis for network traffic monitor system. To overcome these problems, this paper proposes a Multiple-strategies Differential Privacy framework on STF (MDPSTF) for HOHDST network traffic data analysis. MDPSTF comprises three Differential Privacy (DP) mechanisms, i.e., "􀀀 DP, Concentrated DP (CDP), and Local DP (LDP). Furthermore, the theoretical proof of privacy bound is presented. Hence, MDPSTF can provide general data protection for HOHDST network traffic data with high-security promise. We conduct experiments on two real network traffic datasets (Abilene and GEANT). The experimental results show that MDPSTF has high universality on the various degrees of privacy protection demands and high recovery accuracy for the HOHDST network traffic data.

录取期刊：Transactions on Industrial Informatics

标题：《Deep Field-Aware Interaction Machine for Click-Through Rate Prediction》

汇报人：齐高峰

摘要：

Modeling feature interactions is of crucial importance to predict click-through rate (CTR) in industrial recommender systems. Because of great performance and efficiency, the factorization machine (FM) has been a popular approach to learn feature interaction. Recently, several variants of FM are proposed to improve its performance, and they have proven the field information to play an important role. However, feature-length in a field is usually small; we observe that when there are multiple nonzero features within a field, the interaction between fields is not enough to represent the feature interaction between different fields due

to the problem of short feature-length. In this work, we propose a novel neural CTR model named DeepFIM by introducing Fieldaware Interaction Machine (FIM), which provides a layered structure form to describe intrafield and interfield feature interaction, to solve the short-expression problem caused by the short feature-length in the field. Experiments show that our model achieves comparable and even materially better results than the state-of-the-art methods..

录取期刊：Mobile Information Systems