Wenbo Hu 胡文波

Ph.D. Student
Tsinghua University
[Curriculum Vitae]
[Linkedin: wenbohu] [Weibo]

  • [Award] I get the Schlumberger scholarship for the year 2017. Special thanks to Schlumberger Limited. (Oct/17)
  • [Talk] I will be attending IJCAI'2017 soon in Melbourne. Feel free to leave me a message if you want to have a chat with me there. (Aug/17)
  • [Paper] Our paper on "SAM: Semantic Attribute Modulated Language Modeling" is released on arXiv. Find our new model with the interesting lyric generations. (July/17)
  • [Paper] Our latest review paper on large-scale Bayesian learning is published on National Science Review. (May/17)
  • [Paper] Our paper on "Manifold posterior regularized topic model" is accepted to IJCAI'17. (May/17)

About Me

I am a fifth-year Ph.D. candidate of TSAIL Group in Department of Computer Science and Technology in the Tsinghua University. I am jointly supervised by Prof. Bo Zhang and Prof. Jun Zhu. Before that, I received the B. Sc. degree from Department of Applied Mathematics of Xidian University in 2013.

My research interests include machine learning, large-scale inference and their applications in natural language processing.

Publications [Google Scholar]

SAM: Semantic Attribute Modulated Language Modeling
We build an improved RNN-based language model with well-designed semantic attributes. This further motivates us to generate texts with different attribute combinations, with an interesting lyric generation experiment result.
Wenbo Hu, Lifeng Hua, Lei Li, Hang Su, Tian Wang, Ning Chen and Bo Zhang.
arXiv: 1707.00117, 2017.

Semi-supervised Max-margin Topic Models with Manifold Posterior Regularization
We introduce the manifold posterior regularization to the max-margin topic model and present a semi-supervised topic model. With much less document labels, we still learn nice document representation and predict document labels.
Wenbo Hu, Jun Zhu, Hang Su, Jingwei Zhuo, and Bo Zhang.
International Joint Conference on Artificial Intelligence (IJCAI), Melbourne, Australia, 2017.

Big learning with Bayesian Methods
We extensively review the recent advances of large-scale Bayesian learning.
Jun Zhu, Jianfei Chen, Wenbo Hu and Bo Zhang.
National Science Review (NSR), (arXiv:1411.6370), 2017.

Fast Sampling for Bayesian Max-Margin Models
We build a stochastic subgradient MCMC methods for the fast inference of max-margin-regularized posteriors. Experimental results on several Bayesian max-margin models shows that our method is accurate and fast.
Wenbo Hu, Jun Zhu and Bo Zhang.
Expert Systems with Applications (ESWA), (arXiv:1504.07107), 2017

Recent Advances in Bayesian Machine Learning.
We review the recent advances of Bayesian machine learning.
Jun Zhu and Wenbo Hu.
Journal of Computer Science and Development, 2015. (In Chinese)

Honors and Awards

Best Poster Award at AEARU-CSWT Workshop, 2015
Ten Distinguished Undergraduate Students of Xidian University, 2013
Finalist of Interdisciplinary Contest in Modeling(Top 1%), 2012

Last updated on Sep 2017.