University of Electronic Science and Technology of China

About: University of Electronic Science and Technology of China is a education organization based out in Chengdu, China. It is known for research contribution in the topics: Antenna (radio) & Dielectric. The organization has 50594 authors who have published 58502 publications receiving 711188 citations. The organization is also known as: UESTC.

Gate Decorator: Global Filter Pruning Method for Accelerating Deep Convolutional Neural Networks

Abstract: Filter pruning is one of the most effective ways to accelerate and compress convolutional neural networks (CNNs). In this work, we propose a global filter pruning algorithm called Gate Decorator, which transforms a vanilla CNN module by multiplying its output by the channel-wise scaling factors (i.e. gate). When the scaling factor is set to zero, it is equivalent to removing the corresponding filter. We use Taylor expansion to estimate the change in the loss function caused by setting the scaling factor to zero and use the estimation for the global filter importance ranking. Then we prune the network by removing those unimportant filters. After pruning, we merge all the scaling factors into its original module, so no special operations or structures are introduced. Moreover, we propose an iterative pruning framework called Tick-Tock to improve pruning accuracy. The extensive experiments demonstrate the effectiveness of our approaches. For example, we achieve the state-of-the-art pruning ratio on ResNet-56 by reducing 70% FLOPs without noticeable loss in accuracy. For ResNet-50 on ImageNet, our pruned model with 40% FLOPs reduction outperforms the baseline model by 0.31% in top-1 accuracy. Various datasets are used, including CIFAR-10, CIFAR-100, CUB-200, ImageNet ILSVRC-12 and PASCAL VOC 2011.

DeepLiDAR: Deep Surface Normal Guided Depth Prediction for Outdoor Scene from Sparse LiDAR Data and Single Color Image

Abstract: In this paper, we propose a deep learning architecture that produces accurate dense depth for the outdoor scene from a single color image and a sparse depth. Inspired by the indoor depth completion, our network estimates surface normals as the intermediate representation to produce dense depth, and can be trained end-to-end. With a modified encoder-decoder structure, our network effectively fuses the dense color image and the sparse LiDAR depth. To address outdoor specific challenges, our network predicts a confidence mask to handle mixed LiDAR signals near foreground boundaries due to occlusion, and combines estimates from the color image and surface normals with learned attention maps to improve the depth accuracy especially for distant areas. Extensive experiments demonstrate that our model improves upon the state-of-the-art performance on KITTI depth completion benchmark. Ablation study shows the positive impact of each model components to the final performance, and comprehensive analysis shows that our model generalizes well to the input with higher sparsity or from indoor scenes.

Graphene-coated microfiber Bragg grating for high-sensitivity gas sensing

Abstract: A graphene coated microfiber Bragg grating (GMFBG) for gas sensing is reported in this Letter. Taking advantage of the surface field enhancement and gas absorption of a GMFBG, we demonstrate an ultrasensitive approach to detect the concentration of chemical gas. The obtained sensitivities are 0.2 and 0.5 ppm for NH3 and xylene gas, respectively, which are tens of times higher than that of a GMFBG without graphene for tiny gas concentration change detection. Experimental results indicate that the GMFBG-based NH3 gas sensor has fast response due to its highly compact structure. Such a miniature fiber-optic element may find applications in high sensitivity gas sensing and trace analysis.

Ambient N2 fixation to NH3 electrocatalyzed by a spinel Fe3O4 nanorod.

Abstract: Industrially, NH3 is mainly produced via the Haber-Bosch process which is not only energy-consuming but emits a large amount of CO2. Electrochemical reduction is regarded as an environmentally-benign alternative for sustainable NH3 synthesis, and its efficiency heavily depends on the identification of Earth-abundant catalysts with high activity for the N2 reduction reaction (NRR). In this work, we report that a spinel Fe3O4 nanorod on a Ti mesh (Fe3O4/Ti) acts as an efficient and durable NRR electrocatalyst under ambient conditions. When tested in 0.1 M Na2SO4, such Fe3O4/Ti achieves a high faradaic efficiency of 2.6% and a NH3 yield 5.6 × 10-11 mol s-1 cm-2 and at -0.4 V vs. a reversible hydrogen electrode.

Deep Deterministic Policy Gradient (DDPG)-Based Energy Harvesting Wireless Communications

Abstract: To overcome the difficulties of charging the wireless sensors in the wild with conventional energy supply, more and more researchers have focused on the sensor networks with renewable generations. Considering the uncertainty of the renewable generations, an effective energy management strategy is necessary for the sensors. In this paper, we propose a novel energy management algorithm based on the reinforcement learning. By utilizing deep deterministic policy gradient (DDPG), the proposed algorithm is applicable for the continuous states and realizes the continuous energy management. We also propose a state normalization algorithm to help the neural network initialize and learn. With only one day’s real solar data and the simulative channel data for training, the proposed algorithm shows excellent performance in the validation with about 800 days length of real solar data. Compared with the state-of-the-art algorithms, the proposed algorithm achieves better performance in terms of long-term average net bit rate.