科研成果 by Type: Conference Paper

2021
Yang E-Y, Jia T, Brooks D, Wei G-Y. FlexACC: A programmable accelerator with application-specific ISA for flexible deep neural network inference, in International Conference on Application-specific Systems, Architectures and Processors (ASAP).; 2021.
Li S, Wang C, Xie G. Formation control of multiple nonholonomic vehicles with local measurements in 3D space, in 2021 60th IEEE Conference on Decision and Control (CDC). IEEE; 2021:7112–7117. 访问链接
Zhao H, Shi Y, Tong X, Wen J, Ying X, Zha H. G-FAN: graph-based feature aggregation network for video face recognition, in International Conference on Pattern Recognition. IEEE; 2021:1672–1678.
Yan P, Schroeder R. Globalization and anti-globalization, media trust, and populism: A comparative study of the US and Germany., in The 71st Annual International Communication Association (ICA) Conference.; 2021.
Zhang C, Gao X, Ma L, Wang Y, Wang J, Tang W. GRASP: generic framework for health status representation learning based on incorporating knowledge from similar patients, in Proceedings of the AAAI conference on artificial intelligence.Vol 35.; 2021:715–723.
Xu J, Niu Y, Wu X, Qu T. Higher order ambisonics compression method based on independent component analysis, in Audio Engineering Society Convention 150.; 2021:10456.
Zheng Y+, Zheng L+, Yu Z*, Shi B, Tian YH, Huang T. High-Speed Image Reconstruction Through Short-Term Plasticity for Spiking Cameras, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).; 2021:6358-6367. PDFAbstract
Fovea, located in the centre of the retina, is specialized for high-acuity vision. Mimicking the sampling mechanism of the fovea, a retina-inspired camera, named spiking camera, is developed to record the external information with a sampling rate of 40,000 Hz, and outputs asynchronous binary spike streams. Although the temporal resolution of visual information is improved, how to reconstruct the scenes is still a challenging problem. In this paper, we present a novel high-speed image reconstruction model through the short-term plasticity (STP) mechanism of the brain. We derive the relationship between postsynaptic potential regulated by STP and the firing frequency of each pixel. By setting up the STP model at each pixel of the spiking camera, we can infer the scene radiance with the temporal regularity of the spike stream. Moreover, we show that STP can be used to distinguish the static and motion areas and further enhance the reconstruction results. The experimental results show that our methods achieve state-of-the-art performance in both image quality and computing time.
Tao Y, Zhang Z. Hima: A fast and scalable history-based memory access engine for differentiable neural computer, in MICRO-54: 54th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO).; 2021:845–856. 访问链接
Liu Z, Li* S. Impact of Gain-Loss Message Framing on Bedtime Procrastination of College Students: From the Perspective of the Powerful Effect Theory, in 2021 ASIS&T Conference. Salt Lake, USA; 2021.
Liu X, Gui L, Tang F, Jin Y, Chen K, Lang L. Impact of Zooplankton on Underwater Wireless Optical Channel Transmission, in IEEE International Conference on Microwave and Millimeter Wave Technology (ICMMT). Nanjing, China: IEEE; 2021. 访问链接Abstract
Underwater wireless optical communication has attracted widespread attention due to its advantages of high bandwidth and low delay. Seawater environment contains different substances, which will affect the received intensity and time delay of communication. This paper proposes an adaptive Monte-Carlo method to analyze the impact of zooplankton on the received intensity, receiving time, spatial distribution of energy, transmission distance and misalignment. According to the simulation results, when seawater contains more zooplankton, the received intensity is weaker, the receiving time is longer, and the energy is more dispersed.
Yang J, Shi Y, Tong X, Wang R, Chen T, Ying X. Improving Knowledge Graph Embedding Using Affine Transformations of Entities Corresponding to Each Relation, in Findings of the Association for Computational Linguistics: EMNLP 2021, Virtual Event / Punta Cana, Dominican Republic, 16-20 November, 2021. Association for Computational Linguistics; 2021:508–517. 访问链接
Fang W, Yu Z*, Chen Y, Masquelier T, Huang T, Tian YH*. Incorporating Learnable Membrane Time Constant to Enhance Learning of Spiking Neural Networks, in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV).; 2021:2661-2671. PDFAbstract
Spiking Neural Networks (SNNs) have attracted enormous research interest due to temporal information processing capability, low power consumption, and high biological plausibility. However, the formulation of efficient and high-performance learning algorithms for SNNs is still challenging. Most existing learning methods learn weights only, and require manual tuning of the membrane-related parameters that determine the dynamics of a single spiking neuron. These parameters are typically chosen to be the same for all neurons, which limits the diversity of neurons and thus the expressiveness of the resulting SNNs. In this paper, we take inspiration from the observation that membrane-related parameters are different across brain regions, and propose a training algorithm that is capable of learning not only the synaptic weights but also the membrane time constants of SNNs. We show that incorporating learnable membrane time constants can make the network less sensitive to initial values and can speed up learning. In addition, we reevaluate the pooling methods in SNNs and find that max-pooling will not lead to significant information loss and have the advantage of low computation cost and binary compatibility. We evaluate the proposed method for image classification tasks on both traditional static MNIST, Fashion-MNIST, CIFAR-10 datasets, and neuromorphic N-MNIST, CIFAR10-DVS, DVS128 Gesture datasets. The experiment results show that the proposed method outperforms the state-of-the-art accuracy on nearly all datasets, using fewer time-steps. Our codes are available at https://github.com/fangwei123456/Parametric-Leaky-Integrate-and-Fire-Spiking-Neuron.
Zhang M, Guan T, Chen L, Fu T, Su D, Qu T. Individualized HRTF-based Binaural Renderer for Higher-Order Ambisonics, in Audio Engineering Society Convention 150.; 2021:10454.
Li X, Li J, Sun X, Fan C, Zhang T, Wu F, Meng Y, Zhang J. $ k $ Folden: $ k $-Fold Ensemble for Out-Of-Distribution Detection, in Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing.; 2021.
Chun Fan, Jiwei Li TZXAFWYM, Sun X. Layer-wise Model Pruning based on Mutual Information, in Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. 3090; 2021:3079.
Liang Z, Tang K, Dong J, Li Q, Zhou Y, Zhu R, Wu Y, Han D, HUANG R. A Novel High-Endurance FeFET Memory Device Based on ZrO2 Anti-Ferroelectric and IGZO Channel, in 2021 IEEE International Electron Devices Meeting (IEDM).; 2021:17.3.1-17.3.4.
Zhang Y, Li J, Lei Y, Yang T, Li Z, Zhang G, Cui B. On-Off Sketch: A Fast and Accurate Sketch on Persistence, in Proceedings of the VLDB Endowment. VLDB Endowment; 2021.
Ding J, Yu Z*, Tian YH*, Huang T. Optimal ANN-SNN Conversion for Fast and Accurate Inference in Deep Spiking Neural Networks, in Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence (IJCAI).; 2021:2328-2336. PDFAbstract
Spiking Neural Networks (SNNs), as bio-inspired energy-efficient neural networks, have attracted great attentions from researchers and industry. The most efficient way to train deep SNNs is through ANN-SNN conversion. However, the conversion usually suffers from accuracy loss and long inference time, which impede the practical application of SNN. In this paper, we theoretically analyze ANN-SNN conversion and derive sufficient conditions of the optimal conversion. To better correlate ANN-SNN and get greater accuracy, we propose Rate Norm Layer to replace the ReLU activation function in source ANN training, enabling direct conversion from a trained ANN to an SNN. Moreover, we propose an optimal fit curve to quantify the fit between the activation value of source ANN and the actual firing rate of target SNN. We show that the inference time can be reduced by optimizing the upper bound of the fit curve in the revised ANN to achieve fast inference. Our theory can explain the existing work on fast reasoning and get better results. The experimental results show that the proposed method achieves near loss-less conversion with VGG-16, PreActResNet-18, and deeper structures. Moreover, it can reach 8.6× faster reasoning performance under 0.265× energy consumption of the typical method. The code is available at https://github.com/DingJianhao/OptSNNConvertion-RNL-RIL.
Gui X, Gui L, Leng K, Chen X, Liu L, Tang F, Chen K, Lang L. Phase Gradient Metasurface Wall Structure For Antenna Array Decoupling, in IEEE International Conference on Microwave and Millimeter Wave Technology (ICMMT). Nanjing, China: IEEE; 2021. 访问链接Abstract
A gradient metasurface wall structure operating in Ka-band is proposed. To reduce mutual coupling, side metallic walls with a designed metasurface arrays are incorporated into both sides of the unit antenna in the antenna array. These walls can also improve the performance of antenna based on the anomalous reflection. Simulation and experimental results show that the coupling coefficient reduces from -30 dB to -40 dB at the operating frequency. The 3dB beam width of the antenna is broadened from 35° to 38° in H plane and from 23.4° to 31.5° in E plane, due to the characteristic of controllable reflection direction. Because of the freedom in controlling the reflected waves and the high isolation, this gradient metasurface wall structure is of important application values in Millimeter-wave imaging system and MIMO system.
Shi Y, Tong X, Wen J, Zhao H, Ying X, Zha H. Position-aware and symmetry enhanced GAN for radial distortion correction, in International Conference on Pattern Recognition. IEEE; 2021:1701–1708.

Pages