Haotong Qin (秦浩桐)

I was born on July, 1997. Currently, I am a PhD candidate (2019.09-) in the State Key Laboratory of Software Development Environment (SKLSDE) and Shen Yuan Honors College at Beihang University, supervised by Prof. Wei Li and Prof. Xianglong Liu. I obtained my BSc degree in Computer Science and Engineering (Summa Cum Laude) from Beihang University in 2019.

Now I am a research intern (2021.10-) of Bytedance AI Lab, and I was interned at Tencent WXG in 2020. In my undergraduate study, I interned at the Speech Group of Microsoft Research Asia.

Email: qinhaotong@buaa.edu.cn / htqin@outlook.com

CV / Github / Linkedin / Google Scholar / 知乎专栏

Research

I'm interested in hardware-friendly deep learning. And my research goal is to enable state-of-the-art neural network models to be deployed on resource-limited hardware, which includes the compression and acceleration for multiple architectures, and the flexible and efficient deployment on multiple hardware.

My research focus is mainly on:
  • Network binarization and quantization
  • Efficient neural architecture design
  • Hardware implementation of compact network
News

[Call for Papers] I am co-organizing the Workshop on 1st International Workshop on Practical Deep Learning in the Wild (PracticalDL-22) at AAAI 2022. Please submit your papers and win the prizes!

[2021.09.20] I obtain China National Scholarship (Top2%, the 2nd time).

[2021.07.23] One co-authored paper for object detection is accepted by ICCV 2021.

[2021.05.17] I obtain Huawei Scholarship (Top1%).

[2021.05.05] I am selected in Doctoral Consortium IJCAI 2021 (7 people from Mainland China).

[2021.03.01] One co-first-authored oral paper for data-free quantization is accepted by CVPR 2021.

[2021.01.13] One first-authored paper for PointNet binarization is accepted by ICLR 2021.

[2020.09.20] I obtain China National Scholarship (Top2%).

Survey
PontTuset

Binary Neural Network: A Survey [PDF]
Haotong Qin, Ruihao Gong, Xianglong Liu, Xiao Bai, Jingkuan Song, Nicu Sebe
Pattern Recognition (PR), 2020
arXiv / News: (PaperWeekly, 机器之心) / Code

In this paper, we presented a comprehensive survey of these algorithms. We also investigated other practical aspects of binary neural networks such as the hardware-friendly design and the training tricks. Then, we gave the evaluation and discussions on different tasks. Finally, the challenges that may be faced in future research were prospected.

Selected Papers | All Publications
PontTuset

Distribution-sensitive Information Retention for Accurate Binary Neural Network [PDF]
Haotong Qin, Xiangguo Zhang, Ruihao Gong, Yifu Ding, Yi Xu, Xianglong Liu
arXiv 2021

We present a novel Distribution-sensitive Information Retention Network (DIR-Net) to retain the information of the forward activations and backward gradients, which improves BNNs by distribution-sensitive optimization without increasing the overhead in the inference process.

PontTuset

Diverse Sample Generation: Pushing the Limit of Data-free Quantization [PDF]
Haotong Qin, Yifu Ding, Xiangguo Zhang, Aoyu Li, Jiakai Wang, Xianglong Liu, Jiwen Lu
arXiv 2021 / Code

This paper presents a generic Diverse Sample Generation (DSG) scheme for the generative data-free post-training quantization and quantization-aware training, to mitigate the detrimental homogenization.

PontTuset

BiPointNet: Binary Neural Network for Point Clouds [PDF]
Haotong Qin*, Zhongang Cai*, Mingyuan Zhang*, Yifu Ding, Haiyu Zhao, Shuai Yi, Xianglong Liu, Hao Su (*equal contribution)
International Conference on Learning Representations (ICLR), 2021
arXiv / Project / News: (量子位, 商汤学术) / Code

We presented BiPointNet, the first model binarization approach for efficient deep learning on point clouds. BiPointNet gave an impressive 14.7× speedup and 18.9× storage saving on real-world resource-constrained devices.

PontTuset

Diversifying Sample Generation for Accurate Data-Free Quantization [PDF]
Xiangguo Zhang*, Haotong Qin*, Yifu Ding, Ruihao Gong, Qinghua Yan, Renshuai Tao, Yuhang Li, Fengwei Yu, Xianglong Liu (*equal contribution)
IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Oral, 2021
arXiv / News: (量子位, 商汤学术)

We proposed Diverse Sample Generation (DSG) scheme to mitigate the adverse effects caused by homogenization in data-free quantization, which obtained significant improvements over various networks and quantization methods.

PontTuset

Forward and Backward Information Retention for Accurate Binary Neural Networks [PDF]
Haotong Qin, Ruihao Gong, Xianglong Liu, Mingzhu Shen, Ziran Wei, Fengwei Yu, Jingkuan Song
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020
arXiv / News: (机器之心, 商汤学术, CVer, AI科技大本营) / Code

We proposed a novel Information Retention Network (IR-Net) to retain the information that consists in the forward activations and backward gradients, and we were the first to implement and report 1-bit BNN speed on edge devices.

Honors

[2021.09]    China National Scholarship (the 2nd time)

[2021.05]    Huawei Scholarship

[2020.09]    China National Scholarship

[2019.10]    Enrolled in the Tencent Rhino-Bird Elite Training Program (51 people worldwide)

[2019.10]    Enrolled in the Shen Yuan Honors College at Beihang University

[2019.04]    The ICPC China National Invitational Contest (Nanchang, China)   Gold Medal

[2018.03]    The ACM-ICPC Chinese Collegiate Programming Contest (Shizuishan, China)   Gold Medal

[2016.07]    The International Concert of Chinese Folk Music (Kobe, Japan)   Gold Medal

Talks & Academic Services

[2021.07] I am invited to talk about Network Quantization in Multiple Scenarios at J Ventures (将门创投). [Video]

[2021.06] I am invited to talk about Data-free Quantization at Zhidx (智东西公开课). [Video]

[2021.05] I am invited to present our DSG (CVPR 2021 oral) at MSRA Tech Talk. [Slides]

[2020.04] I am invited to present our IR-Net (CVPR 2020) and survey paper at JD AI Research. [Slides]

[2020-] I regularly review papers for top-tier conferences and journals in machine learning and computer vision.

Teaching

[Fall 2020]    Teaching Assistant in Machine Learning (Beihang University).

About Me

In my free time, I like playing Chinese folk music, especially string music (Erhu, Zhonghu, etc.). In fact, I am almost a professional Erhu performer. I have studied Erhu supervised by Prof. Zaili Tian, Prof. Yang Gao, and Prof. Qingfu Zhu. I was the vice-president of the Beihang Folk Music Orchestra, here are some of the performance videos of our orchestra [Tencent Video].