Hao Zhang (张皓)

 

Hao Zhang (Chinese: 张皓; born Feb. 1994) 

Researcher

Tencent Video

PCG, Tencent

Email (change # to @): zhangh0214#gmail.com

[知乎] [Google Scholar] [GitHub] [LinkedIn]

 

 

Biography

See the source image

腾讯视频Logo

 

 

Since 2021, Hao is working as a researcher in Tencent Video. Tencent Video is affiliated with the PCG (Platform & Content Group), Tencent. Tencent Video was officially launched in 2011. Until 2020, it has 123 million paid members. The average daily usage time per user is 110 minutes.

See the source image

From 2019 to 2021, Hao was working as a researcher in YouTu Lab, founded by Yunsheng Wu and professor Jiaya Jia (IEEE Fellow). The current leader is Yunsheng Wu. YouTu Lab is affiliated with the CSIG (Cloud and Smart Industries Group), Tencent. YouTu Lab is one of the three AI research labs in Tencent, which focuses on computer vision. This team is sound for intelligent video understanding, review, editing, etc. This team is severing several internal products in Tencent, such as WeChat Top Stories (微信看一看), Tencent Kandian (腾讯看点), and Tencent WeSee (腾讯微视).

 

Before that, Hao received his M.Sc.'s degree at LAMDA Group, which is led by professor Zhi-Hua Zhou (foreign member of the Academy of Europe, ACM/AAAI/AAAS/IEEE/IAPR Fellow). His adviser was professor Jianxin Wu (member of the Thousand Talents Plan). LAMDA (Learning And Mining from DatA) is affiliated with the National Key Laboratory for Novel Software Technology, Department of Computer Science and Technology and the School of Artificial Intelligence, Nanjing University (NJU), China. He received his B.Sc.'s degree in Nanjing University in 2016. His GPA ranked No.1 during his undergraduate studies. In the same year, he was admitted to study for a M.Sc.'s degree without entrance examination. He was a member of Excellent Engineer Training Program, from Ministry of Education of China. This link provides a short interview of Hao.

 

 

Besides, he has been a member of Communist Party of China since 2013. He is a columnist/writer of serval medias, such as 机器之心, 新智元, AI科技评论, 人工智能头条, SIGAI, and AI100.

 

 

Research Interests

Hao's current research interests mainly include machine learning, computer vision, especially on deep learning and visual recognition. He is working on exploiting convolutional features in both supervised and unsupervised ways to improve the efficiency of convolutional neural networks.

Besides, he is particularly interested in linear algebra and its applications. Linear algebra is closely related to real-world applications, such as linear systems (Ax = b), dynamic systems (uk+1 = Auk or du(t)/dt = Au(t)), optimization problems (argminu f(u)), linear transformations (T(u) = Au), etc. Please refer to this page for further discussions.

 

 

Publications

吴建鑫著. 罗建豪, 张皓. 吴建鑫审校. 模式识别. 机械工业出版社, ISBN: 9787111643890, 1-368, 2020. [简介和勘误] [京东链接] [当当链接]

https://img10.360buyimg.com/n1/jfs/t1/198016/26/9608/134081/614c17c9E0e3e3b4a/068b800b2a1ae48c.jpg

张皓编著. 深度学习视频理解. 电子工业出版社, ISBN: 9787121419805, 1-314, 2021. [简介和勘误] [京东链接] [当当链接]

图示, 示意图

描述已自动生成

张皓. 手把手带你学懂 SVM. 1-118, 2022. [简介和勘误] [PDF下载链接1] [PDF下载链接2]

Jian-Hao Luo, Hao Zhang, Hong-Yu Zhou, Chen-Wei Xie, Jianxin Wu, and Weiyao Lin. ThiNet: Pruning CNN filters for a thinner net. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 41(10): 2525-2538, 2019. [pdf] [code]

 

 

Chen-Lin Zhang, Hao Zhang, Xiu-Shen Wei, and Jianxin Wu. Deep bimodal regression for apparent personality analysis. In Proceedings of the 14th European Conference on Computer Vision (ECCV'16) Workshops, LNCS 9915, pages 311-324, 2016. [pdf] [project page] [slides] [code]

 

 

Xiu-Shen Wei, Chen-Lin Zhang, Hao Zhang, and Jianxin Wu. Deep bimodal regression of apparent personality traits from short video sequences. IEEE Transactions on Affective Computing (TAC), 9(3): 303-315, 2018. [pdf] [project page] [code]

 

 

Hao Zhang and Jianxin Wu. A survey on unsupervised image retrieval using deep features. Journal of Computer Research and Development (CRAD), 55(9): 1829-1842, 2018. [pdf] (in Chinese)

 

 

 

Hao Zhang and Jianxin Wu. Ensemble max-pooling: Is only the maximum activation useful when pooling. In Proceedings of CCF Conference on Artificial Intelligence (CCFAI), 2017. [pdf] [poster] [spotlight] (acceptance rate: 34.0%) (in Chinese)

 

 

Hao Zhang and Jianxin Wu. Ensemble max-pooling: Is only the maximum activation useful when pooling. Journal of University of Science and Technology of China (JUST), 47(10): 799-807, 2017. [pdf] (in Chinese)

 

 

Awards and Honors

Four-Star Distinguished Staff (Twice)

Tencent

2019, 2021

Company-level Technology Breakthrough Award

Tencent

2019

YouTu “Willing to Share” Award

Tencent

2020

YouTu YouTech Preacher

Tencent

2020

Winner of the Apparent Personality Analysis Competition

ECCV

2016

National Scholarship

Ministry of Education of China

2013

Member of Excellent Engineer Training Program

Ministry of Education of China

2013--2016

Excellent All-round Student

Jiangsu Provincial Department of Education

2015

Top-grade in Jiangsu Provincial Undergraduate Electronics Design Contest

Jiangsu Committee of the National Undergraduate Electronics Design Contest

2014

Top-grade "Red Sun" Scholarship (only 20/12,000 undergraduate students in NJU achieved per year)

Nanjing Red Sun CO., LTD

2014

Top-grade Graduate Research Scholarship

Nanjing University

2016--2018

Excellent Students

Nanjing University

2014

Excellent Cadre of Students

Nanjing University

2013

Excellent Undergraduate Student

Nanjing University

2016

People's Scholarship Speciality Specialization

Nanjing University

2015

People's Scholarship Social Work Specialization

Nanjing University

2014

Second Class Xingquan Responsibility Scholarship

Nanjing University

2015

Excellent League Member (twice)

Youth League Committee of Nanjing University

2014, 2015

Excellent Student in Summer Social Practice

Youth League Committee of Nanjing University

2013

 

 

Articles

深度学习

n   三次简化一张图:一招理解LSTM/GRU门控机制 [link] (repost on 机器之心, AI100, 搜狐, 网易, 机器学习研究组订阅)

n   深度学习基础及数学原理 [pdf] [slides]

n   深度学习基础(基本概念、优化算法、初始化、正则化等) [link] (repost on 新智元, 搜狐)

n   计算机视觉四大基本任务(分类、定位、检测、分割) [link] (repost on 新智元, 搜狐, 新机器视觉, 矩视智能, 深度学习自然语言处理, 小白学视觉, etc)

n   计算机视觉其他应用(网络压缩、视觉问答、可视化、风格迁移等) [link] (repost on 新智元, 人工智能头条, 人工智能学家, 前沿技术研究, 中国指挥与控制学会, etc)

n   双线性汇合(Bilinear Pooling)在细粒度图像分析及其他领域的进展综述 [link] (repost on SIGAI)

n   深度学习项目实践技巧指南 [link] (repost on SIGAI, AI算法与图像处理)

n   视频理解近期研究进展 [link] (repost on 新智元, 搜狐)

n   动作识别时序汇合(Temporal Pooling)方法介绍 [link] (repost on 人工智能前沿讲习)

n   YouTube推荐系统算法梳理 [link] (repost on 腾讯优图AI开放平台, 深度传送门, 智能推荐系统, 小小挖掘机, etc)

n   工业界视频理解解决方案大汇总 [link] (repost on 人工智能前沿讲习)

 

 

机器学习 (repost on AI科技评论)

n   人工智能科普文:AI的过去和现在 [link]

n   第一章:绪论 [pdf]

n   第二章:学习理论 [pdf]

n   第三章:模型评估 [pdf]

n   第四章:概率方法 [pdf]

n   第五章:支持向量机 [pdf]

n   第六章:集成学习 [pdf]

n   第七章:决策树 [pdf]

n   附录(机器学习数学基础速查手册):线性代数 [pdf], 概率论 [pdf]

n   当你在应用机器学习时你应该想什么 [link] (repost on AI100, 开发者头, 搜狐)

 

 

See the source image

开发工具

n   PyTorch Cookbook (常用代码段整理合集) [link] (repost on 机器之心, AI有道, 雷课, 极市平台, AI, DeeAo大数据, 深度学习这件小事, AI算法修炼营, etc)

n   Python 速运行技巧 [link] (repost on 极市平台, 机器学习算法那些事, 马哥Linux运维)

n   论文格式排版你真的做对了吗? 常用格式及其LaTeX书写方法介绍 [link] (repost on 机器之心, 搜狐, 凤凰网科技, 新浪, 极客头条)

 

 

A simple and efficient implementation of im2col in convolution neural networks [pdf]

 

 

 

 

 

 

在电子,你会被加哪些技能点--电子学院专业课程分析 [pdf]

 

 

 

 

 

 

停机问题(C语言版) [link]

 

 

 

 

Notes

n   Notes on machine learning [pdf]

n   Notes on mathematics for computer science [pdf]

Notes on C++: Concepts and practices [pdf]

 

 

MOOC and Online Courses

Hao is determined to become a lifelong learner. The followings are the courses he completed.

n   John Guttag. 6.00: Introduction to computer science and programming. MIT, 2011.

n   Dennis Freeman. 6.01: Introduction to electrical engineering and computer science I. MIT, 2011.

n   Erik Demaine and Srinivas Devadas. 6.006: Introduction to algorithms. MIT, 2011.

n   Charles Leiserson and Erik Demaine. 6.046: Introduction to algorithms. MIT, 2005.

n   John Tsitsiklis. 6.041: Probabilistic systems analysis and applied probability. MIT, 2013.

n   Tom Leighton and Marten van Dijk. 6.042: Mathematics for computer science. MIT, 2010.

n   Albert Meyer and Adam Chlipala. 6.042: Mathematics for computer science. MIT, 2015.

n   Erik Demaine, Srinivas Devadas, and Nancy Lynch. 6.046: Design and analysis of algorithms. MIT, 2015.

n   Sarina Canelake. 6.189: A Gentle introduction to programming using Python. MIT, 2011.

n   David Jerison. 18.01: Single variable calculus. MIT, 2010.

n   Denis Auroux. 18.02: Multi-variable calculus. MIT, 2010.

n   Jeremy Orloff and Jonathan Bloom. 18.05: Introduction to probability and statistics. MIT, 2014.

n   Gilbert Strang. 18.06: Linear algebra. MIT, 2011.

 

n   Chris Manning and Richard Socher. CS224n: Natural language processing with deep learning. Stanford.

n   Andrew Ng. CS229: Machine learning. Stanford.

n   Anand Avati. CS229: Machine learning. Stanford.

n   Andrew Ng and Kian Katanforoosh. CS230: Deep learning. Stanford.

n   Fei-Fei Li, Justin Johnson, and Serena Yeung. CS231n: Convolutional neural networks for visual recognition, Stanford.

n   Div Garg, Chetanya Rastogi, Advay Pal, and Chris Manning. CS25: Transformers united, Stanford.

n   Andrew Ng et al. UFLDL: Deep learning tutorial. Stanford.

 

n   Michael Fitzpatrick.  Introduction to programming with MATLAB. Vanderbilt University. [Grade: 100%] [Statement of Accomplishment (with distinction)]

n   Hsuan-Tien Lin. Machine learning foundations. National Taiwan University. [Grade: 96.4%] [Certificate (with distinction)]

n   Hsuan-Tien Lin. Machine learning techniques. National Taiwan University. [Grade: 96.0%] [Certificate (with distinction)]

n   Andrew Ng. Machine learning. Stanford. [Grade: 100%] [Older version]

n   Andrew Ng. Neural networks and deep learning. deeplearning.ai, 2017.

n   Andrew Ng. Improving deep neural networks: Hyperparameter tuning, regularization and optimization. deeplearning.ai, 2017.

n   Andrew Ng. Structuring machine learning projects. deeplearning.ai, 2017.

n   Andrew Ng. Convolutional neural networks. deeplearning.ai, 2017.

n   Andrew Ng. AI for everyone. deeplearning.ai, 2019

n   Guillermo Sapiro. Image and video processing from Mars to Hollywood with a stop at the hospital. Duke University. [Solution Code]

n   陈昌凯. 心理学与生活. Nanjing University.

n   吕世浩. 中国古代历史与人物---秦始皇. National Taiwan University.

n   吕世浩. 史記(一). National Taiwan University.


 

 

Book List

The followings are the books Hao has read.

n   Sheldon Axler. Linear algebra done right. Springer, 1997.

n   Stephen Boyd and Lieven Vandenberghe. Introduction to applied linear algebra: Vectors, matrices, and least squares. Cambridge University Press, 2018.

n   Thomas Cormen, Charles Leiserson, Ronald Rivest, and Clifford Stein. Introduction to algorithms (3rd edition). MIT Press, 2009. [Solutions]

n   Allen Downey, Jeffrey Elkner, and Chris Meyers. How to think like a computer scientist: Learning with Python. Green Tea Press, 2002.

n   Allen Downey. Think Python: how to think like a computer scientist. Green Tea Press, 2012. [Online book]

n   Ian Goodfellow, Aaron Courville, and Yoshua Bengio. Deep Learning: Adaptive computation and machine learning series. MIT Press. 2016.

n   Rafael Gonzalez, Richard Woods, and Steven Eddin. Digital image processing. Pearson, 2007.

n   Rafael Gonzalez, Richard Woods, and Steven Eddin. Digital image processing using MATLAB. Gatesmark Publishing, 2009.

n   David C. Lay. Linear Algebra and Its Applications (Fifth Edition). Pearson, 2014.

n   Eric Lehman, Thomson Leighton, and Albert Meyer. Mathematics for computer science (2010 version). MIT, 2010.

n   Eric Lehman, Thomson Leighton, and Albert Meyer. Mathematics for computer science (2017 version). MIT, 2017.

n   Stanley Lippman. Essential C++. Addison-Wesley Longman Publishing Co., Inc., 1999.

n   Stanley Lippman, Josée Lajoie, and Barbara Moo. C++ primer (5th edition). Addison-Wesley Professional, 2012. [Errata] [Solution code]

n   Bradley Miller and David Ranum. Problem solving with algorithms and data structures using Python. Franklin, Beedle & Associates Inc., 2006. [Online book]

n   Andrew Ng. Machine learning yearning. Draft. 2018.

n   Williams Shotts. The linux command line: A complete introduction. No Starch Press, 2012.

n   Gilbert Strang. Introduction to linear algebra (Fourth Edition). Wellesley Cambridge Press, 2009.

n   Gilbert Strang. Linear algebra and its applications (Fourth Edition). Academic Press, 2006.

n   刘金鹏. Linux入门很简单. 清华大学出版社, 2012.

n   吴军. 数学之美. 第二版. 人民邮电出版社, 2014.

n   王世江、鸟哥. 鸟哥的Linux私房菜:基础学习篇. 3. 人民邮电出版社, 2010.

n   周志华. 机器学习. 清华大学出版社, 2016. [勘误修订]

 

 

Reimplementations

Hao has reimplemented several papers, some of which have been open sourced in his GitHub.

n   Yuchi Huang, Xiuyu Sun, Ming Lu, and Ming Xu. Channel-max, channel-drop and stochastic max-pooling. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, pages 9–17, 2015.

n   Yannis Kalantidis, Clayton Mellina, and Simon Osindero. Cross-dimensional weighting for aggregated deep convolutional features. In Proceedings of the European Conference of Computer Vision (ECCV) Workshops, pages 685--701, 2016.

n   Tsung-Yu Lin, Aruni RoyChowdhury, and Subhransu Maji. Bilinear CNN models for fine-grained visual recognition. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 1449--1457, 2015.

n   Xiu-Shen Wei, Chen-Lin Zhang, Yao Li, Chen-Wei Xie, Jianxin Wu, Chunhua Shen, and Zhi-Hua Zhou. Deep descriptor transforming for image co-localization. In Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), pages 3048--3054, 2017.

n   Haibing Wu and Xiaodong Gu. Max-pooling dropout for regularization of convolutional neural networks. In Proceedings of the International Conference on Neural Information Processing (ICNIP), pages 46--54, 2015.

n   Jian Xu, Cunzhao Shi, Cheng-Zuo Qi, Chunheng Wang, and Baihua Xiao. Part-based weighting aggregation of deep convolutional features for image retrieval. In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), pages 7436--7443, 2018.

 

 

Correspondence

https://haomood.github.io/homepage/ (Continue updating)

http://lamda.nju.edu.cn/zhangh/ (Deprecated)