Yang Song


About Me   Research Interests   Invited Talks   Publications   Professional Activities

About Me

I have joined Kuaishou(HK.1024) as the Head of Recommendation since June 2020, overseeing both modeling and data mining teams. I manage and lead 100+ engineers and scientists to develop machine learning models that power Kuaishou's recommendation engines in all major services. We are one of the world's leading short-video content communities and social platforms with 1 billion monthly active users, and still growing at a tremendous speed (Bloomberg, Forbes). My group is hiring engineers and scientists at all levels in Beijing. Feel free to drop me a line if you are interested.

I was a Research&Engineering Manager in the Deep Dialogue group of Machine Intelligence in Google AI, working on building the world's best dialogue engines for Google products using deep learning, NLP and recommender systems. Before that, I was a Senior Software Engineer at Google Seattle working on user modeling and machine learning for Ads.

I spent over 7 years at Microsoft Research before joining Google. I was a Researcher at the Deep Learning Technology Center at Microsoft Research Redmond. Before that, I was a Researcher affiliated with Internet Services Research Center (ISRC)Search Quality & Cyber-Intelligene Lab (SQ-CIL) in MSR Redmond. I joined MSR in 2009 after I got my Ph.D. During my Ph.D, I worked with my advisor C. Lee Giles on the next generation scientific literature search engine CiteSeer, which was once the world most popular search engines for scientific literature [screenshot].

Here is the link to my Google Scholar page.


[new]The first comprehensive book about query understanding for search engines has been published. We co-authored the chapter of query suggestions.

We have recently open-sourced our code that implemented Multi-view DNN and Temporal DSSM in Keras [github link].

Education

Ph.D. (2004- 2008) in Computer Science and EngineeringThe Pennsylvania State University
B.S. (1999-2003) in Computer Science and Technology, Zhejiang University, China

Research Interests

My research includes a broad interests of machine learning-related fields, e.g., text classification, information retrieval, search engine ranking, recommender systems and so on.

Deep Learning: Recently, I have been focusing on using deep learning for recommendation, ranking and personalization.

Online User Behavioral Genome Sequencing: my recent research has focused on user behavioral analysis on search engines, mobile devices and other online services in order to create high quality user profiles and better tailor online services to meet the user needs. Our recent WWW 2015 DNN paper and  WWW2014 Demo reveals the tip of the iceberg from this project.

From search session to user tasks: In the past three years, I have been leading a project named Search TrailBlazer that aims at redefining search sessions with tasks. Check out our latest project status and code for scientific uses.

Social tagging Recommendation: I've put up a page regarding my research on recommender systems for social bookmarking.

I've been lucky to have worked with many smart students for their summer internships when I was at MSR.

 

Invited Talks & Media Coverages

Publications

Book Chapters

Journal Publications

Referred Conference Proceedings

(*) are students I mentored for their summer internships at Microsoft Research and Google Research.

2022

  • Yu Tian, Jianxin Chang, Yanan Niu, Yang Song and Chenliang Li. When Multi-Level Meets Multi-Interest: A Multi-Grained Neural Model for Sequential Recommendation, in SIGIR 2022.
  • Zihua Si, Xueran Han, Xiao Zhang, Jun Xu, Yue Yin, Yang Song and Ji-Rong Wen, A Model-Agnostic Causal Learning Framework for Recommendation using Search Data, in TheWebConf(WWW) 2022.
  • Yu Zheng, Chen Gao, Jianxin Chang, Yanan Niu, Yang Song, Depeng Jin and Yong Li, Disentangling Long and Short-Term Interests for Recommendation, in TheWebConf(WWW) 2022.
  • Tiankai Gu, Chaokun Wang, Cheng Wu, Yunkai Lou, Jingcao Xu, Changping Wang, Kai Xu, Can Ye and Yang Song, HybridGNN: Learning Hybrid Representation for recommendation in Multiplex Heterogeneous Networks, in ICDE 2022.

2021

  • Jianxin Chang, Chen Gao, Yu Zheng, Yiqun Hui, Yanan Niu, Yang Song, Depeng Jin and Yong Li, Sequential Recommendation with Graph Convolutional Networks, in SIGIR 2021.
  • Xiaoxue Zang, Lijuan Liu, Maria Wang, Yang Song, Hao Zhang and Jindong Chen, PhotoChat: A Human-Human Dialogue Dataset With Photo Sharing Behavior For Joint Image-Text Modeling, in ACL 2021 (main conference).
  • Yiyu Liu, Qian Liu, Yu Tian, Changping Wang, Yanan Niu, Yang Song and Chenliang Li, Concept-Aware Denoising Graph Neural Network for Micro-Video Recommendation, in CIKM 2021.
  • Xinying Song, Alex Salcianu, Yang Song, Dave Dopson and Denny Zhou, Fast WordPiece Tokenization, in EMNLP 2021 (main conference).
  • Sanqiang Zhao*, Raghav Gupta, Yang Song and Denny Zhou, Extremely Small BERT Models from Mixed-Vocabulary Training, in EACL 2021.

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Contact Me:

EMail: ys at sonyis dot me
Kuaishou Technology,
6 Shangdi West Road Haidian District
Beijing, 100085 China
Google Research
747 6th St
Kirkland, WA 98033
Microsoft Research
One Microsoft Way
Redmond, WA 98052