About

DeepBurning started in 2015 as a neural network accelerator design framework and has been led by Dr. Ying Wang in Institute of Computing Technology (ICT). The project grows rapidly with the increasing popularity of neural network accelerators. Now the project is led by Dr. Ying Wang, Dr. Cheng Liu and Dr. Dawen Xu. It involves 5 Ph.D students and 10 master students working for the open source and deep learning acceleration research.

The project is now supported in part by National Natural Science Foundation of China Under grant No. 61874124 and No. 61902375.

Ph.D candidates in ICT

Ms. Mengdi Wang: Multi-core NPU architecture design and optimization
Mr. Shengwen Liang: In-storage deep learning for un-structural data retrieval
Mr. Weiwei Chen: Network architecture (NAS) search for computer architecture design
Mr. Xiandong Zhao: Low-precision quantization for edge AI
Ms. Xuyi Cai: Memory access optimization with machine learning

Msc. Students in ICT

Ms. Ting Hu: Neural network compiler for customized NPU
Ms. Zhiwei Wan: Neural network compiler for customized NPU
Mr. Chang Si: AI deployment on Zynq with DeepBurning
Mr. Lei He: Light-weight AI in smart SSD
Mr. Changyuan Li: Graph neural network accelerator design
Mr. Yongchen Wang: accelerator design for video and sequence learning

Msc. Students in Hefei University of Technology

Mr. Kaijie Tu (Graduated in 2018): Deconvolution acceleration with DeepBurning
Mr. Cheng Chu: Deformable neural network acceleration with DeepBurning
Mr. Kexin Chu: Resilient neural network search
Ms. Meng He: High-throughput dynamic NN processing acceleration using DeepBurning