北京大学-高能效计算与应用中心高能效计算与应用中心发表的相关论文
Bingzhe Wu, Chaochao Chen, Shiwan Zhao, Cen Chen, Yuan Yao,Guangyu Sun, Li Wang,Xiaolu Zhang,Jun Zhou
national conference on artificial intelligence, (2020)
Bayesian deep learning is recently regarded as an intrinsic way to characterize the weight uncertainty of deep neural networks~(DNNs). Stochastic Gradient Langevin Dynamics~(SGLD) is an effective method to enable Bayesian deep learning on large-scale datasets. Previous theoreti...
Cited by2BibtexViews10Links
1
0
Jiajie Li, Yuze Chi,Jason Cong
FPGA, pp.51-57, (2020)
The domain-specific language (DSL) for image processing, Halide, has generated a lot of interest because of its capability of decoupling algorithms from schedules that allow programmers to search for optimized mappings targeting CPU and GPU. Unfortunately, while the Halide commun...
Cited by1BibtexViews20Links
0
0
Licheng Guo, Jason Lau, Yuze Chi,Jie Wang, Cody Hao Yu,Zhe Chen,Zhiru Zhang,Jason Cong
FPGA, pp.311, (2020)
Designs generated by high-level synthesis (HLS) tools typically achieve a lower frequency compared to manual RTL designs. We study the timing issues in a diverse set of nine realistic HLS designs and observe that in most cases the frequency degradation is related to the signal br...
Cited by1BibtexViews18Links
0
0
Yijin Guan,Guangyu Sun, Zhihang Yuan, Xingchen Li,Ningyi Xu, Shu Chen,Jason Cong,Yuan Xie
IEEE Transactions on Computers, no. 7 (2020): 931-943
We propose an accelerator design, Crane, to employ the load-balancing method to address all types of sparsity irregularities in sparse C ONVOLUTIONAL neural networks
Cited by0BibtexViews25Links
0
0
Atefeh Sohrabizadeh,Jie Wang,Jason Cong
FPGA, pp.133-139, (2020)
The irregularity of recent Convolutional Neural Network (CNN) models such as less data reuse and parallelism due to the extensive network pruning and simplification creates new challenges for FPGA acceleration. Furthermore, without proper optimization, there could be significant ...
Cited by0BibtexViews15Links
0
0
IEEE Transactions on Parallel and Distributed Systems, no. 1 (2020): 64-79
We propose a cost-based iterative control framework able to generate the energy-efficient route, eliminate the conflict, and leverage the multi-core parallel techniques to improve the real-time performance for autonomous control systems
Cited by0BibtexViews12Links
0
0
Bochen Tan, Jason Cong
Recent years have witnessed the fast development of quantum computing. Researchers around the world are eager to run larger and larger quantum algorithms that promise speedups impossible to any classical algorithm. However, the available quantum computers are still volatile and...
Cited by0BibtexViews1Links
0
0
Yue Wu, Purui Wang, Kenuo Xu, Lilei Feng,Chenren Xu
SIGCOMM '20: Annual conference of the ACM Special Interest Group on Data Communication on the applic..., pp.186-197, (2020)
We present how to design the preamble for polarization-based QAM rotation correction, a equalizer for delayed superimposition modulation inter-symbol interference elimination in demodulation with a channel training process dedicated for handling sub-channel heterogeneity from dif...
Cited by0BibtexViews5Links
0
0
Nikola Samardzic, Weikang Qiao, Vaibhav Aggarwal,Mau-Chung Frank Chang, Jason Cong
2020 ACM/IEEE 47th Annual International Symposium on Computer Architecture (ISCA), pp.282-294, (2020)
1) Resource Model Results: The resource utilization results reported by the synthesis tool are within 5% of our resource utilization predictions for all adaptive merge tree we were able to implement on the AWS EC2 F1 instance; that is, all adaptive merge trees such that p ≤ 32 an...
Cited by0BibtexViews1Links
0
0
Wentai Zhang,Ming Jiang,Guojie Luo
FCCM, pp.28-32, (2020)
We evaluate a low-memory general matrix multiplications algorithm for convolutional neural network inference on FPGAs
Cited by0BibtexViews2Links
0
0
Shuang Wen,Guojie Luo
FCCM, pp.172-176, (2020)
We propose an FPGA accelerator for the stateof-the-art tomographic alignment algorithm
Cited by0BibtexViews1Links
0
0
Michael Lo,Zhenman Fang, Jie Wang, Peipei Zhou,Mau-Chung Frank Chang,Jason Cong
FCCM, pp.157-166, (2020)
We presented the first algorithm and hardware co-design to accelerate the Base Quality Score Re-calibration algorithm in Genome Analysis ToolKit version 4
Cited by0BibtexViews14Links
0
0
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, no. 11 (2019): 2072-2085
With the recent advancement of multilayer convolutional neural networks (CNNs) and fully connected networks (FCNs), deep learning has achieved amazing success in many areas, especially in visual content understanding and classification. To improve the performance and energy effic...
Cited by262BibtexViews20Links
0
0