BenchCouncil Challenges

BenchCouncil organizes a series of international AI challenges, consisting of system challenges (on many kinds of chips such as Cambricon, X86, RISC-V), and algorithm challenge on prominent AI tasks. The topics for all challenge tracks are derived from AIBench (Technical Report, Bench18).

2019 AI Challenge Award Recipients

  • RISC-V Track First Prize: Jiageng Yu, Yuxia Miao, Yang Tai (Institute of Software, Chinese Academy of Sciences)
  • RISC-V Track Second Prize: Yangyang Kong (Institute of Information Engineering, Chinese Academy of Sciences)
  • Cambricon Track First Prize: Guangli Li, Xueying Wang, Xiu Ma (ICT, CAS)
  • Cambricon Track Second Prize: Zihan Jiang, Jiansong Li (ICT, CAS)
  • Cambricon Track Second Prize: Yifan Wang, Chen Zeng, Chundian Li (ICT, CAS)
  • Cambricon Track Third Prize: Peng He, Ge Chen, Kai Deng ((ICT, CAS))
  • X86 Track First Prize: Weixin Deng, Jing Wang, Pengyu Wang (Shanghai Jiao Tong University)
  • X86 Track Second Prize: Tianshu Hao (ICT, CAS), Ziping Zheng (Google)
  • X86 Track Second Prize: Maosen Chen (360), Qianyun Chen (Georgia Institute of Technology), Tun Chen (ICT, CAS)
  • X86 Track Third Prize: Yi Liang, Shaokang Zeng, Yande Liang, Kaizhong Chen (Beijing University of Technology)
  • Algorithm Track First Prize: Xingwang Xiong, Xu Wen, Cheng Huang (ICT, CAS)
  • Algorithm Track Second Prize: Tongyan Gong (ICT, CAS), Huiqian Niu (JD.com)
  • Algorithm Track Second Prize: Heming Sun, Xi Xiong (The Ohio State University)

2019 AI Challenge Papers

  • International AI System Challenge based on RISC-V
    • RVTensor: A light-weight neural network inference framework based on the RISC-V architecture [PDF]
    • AIRV: Enabling Deep Learning Inference on RISC-V [PDF]
  • International AI System Challenge based on Cambricon Chip
    • XDN: Towards Efficient Inference of Residual Neural Networks on Cambricon Chips [PDF]
    • Performance Analysis of Cambricon MLU100 [PDF]
    • Improve Image Classification by Convolutional Network on Cambricon [PDF]
    • Exploring the Performance Bound of Cambricon Accelerator in End-to-End Inference Scenario [PDF]
  • International AI System Challenge based on X86 Platform
    • PSL: Exploiting Parallelism, Sparsity and Locality to Accelerate Matrix Factorization on x86 Platforms [PDF]
    • The Implementation and Optimization of Matrix Decomposition Based Collaborative Filtering Task on X86 Platform [PDF]
    • An Efficient Implementation of the ALS-WR Algorithm on x86 CPUs [PDF]
    • Accelerating Parallel ALS for Collaborative Filtering on Hadoop [PDF]
  • International 3D Face Recognition Algorithm Challenge
    • Improving RGB-D face recognition via transfer learning from a pretrained 2D network [PDF]
    • An Implementation of ResNet on the Classification of RGB-D Images [PDF]
    • Utilization of Resnet in RGB-D Facial Recognition Problems [PDF]

2019 Challenges

2019 Challenge Awards: 500,000 CNY in total
Special Award (Only one): 100,000 CNY
The First Prize: 30,000 CNY (one for every competition track)
The Second Prize: 20,000 CNY (two for every competition track)
The Third Prize: 10,000 CNY (three for every competition track)

For more information, please visit the discussion group hosted on Xinxiu---a dedicated science and education communication tool. You need register via your email.

This year, joint with Cambricon, China RISC-V alliance, Sugon, and Intellifusion, BenchCouncil organizes four challenge tracks, including 3 system tracks and 1 algorithm track (http://www.benchcouncil.org/competition/index.html).

  • International AI System Challenge based on RISC-V
  • International AI System Challenge based on Cambricon Chip
  • International AI System Challenge based on X86 Platform
  • International 3D Face Recognition Algorithm Challenge

Challenge Registration: http://www.benchcouncil.org/competition/login.html
If you have registered the discussion group, just use the same account.

More details are available from the challenge manual.

BenchCouncil provides (1) BenchCouncil Testbed, including RISC-V simulator, Cambricon chip, Sugon processor, GPU, and other processors or accelerators, (2) code management system---BenchHub, you need submit the code to BenchHub, (3) communication tool---Xinxiu, we will share and update information in the discussion groups hosted on Xinxiu, (4) presentation session---on BenchCouncil International Symposium on Benchmarking, Measuring and Optimizing (Bench) (Bench 19, Nov 14-16, 2019 @ Denver, Colorado, USA, the same place, priori to SC 19).