Mingzhe Wang

I am a Ph.D. student in the Department of Computer Science at Princeton University, where I work with Prof. Jia Deng in the Princeton Vision & Learning Lab.

I received my master’s degree from the University of Michigan, working together with Prof. Jia Deng and Prof. Rada Mihalcea. I received my bachelor’s degree from Peking University, working with Prof. Ming Zhang.

Email  /  Google Scholar  /  Github

profile photo
Research

I am interested in automated reasoning, machine learning, and artificial intelligence in general. My current research focuses on building machine learning techniques to advance automated theorem proving. Prior to that, I worked on joint language and vision recognition and graph representation learning.

Publications
Learning to Prove Theorems by Learning to Generate Theorems
Mingzhe Wang, Jia Deng
arxiv:2002.07019
[paper]

We present MetaGen, a neural theorem generator to generate synthetic theorems for the purpose of training the neural theorem prover.

Speaker Naming in Movies
Mahmoud Azab, Mingzhe Wang, Max Smith, Noriyuki Kojima, Jia Deng, Rada Mihalcea
Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL), 2018.
[paper][dataset]

We propose a new model for speaker naming in movies that leverages visual, textual, and acoustic modalities in an unified optimization framework.

Premise Selection for Theorem Proving by Deep Graph Embedding
Mingzhe Wang, Yihe Tang, Jian Wang, Jia Deng
Neural Information Processing Systems (NeurIPS), 2017.
[paper][code]

We propose a graph embedding-based approach to the problem of premise selection: selecting mathematical statements relevant for proving a given conjecture.

Structured Matching for Phrase Localization
Mingzhe Wang, Mahmoud Azab, Noriyuki Kojima, Rada Mihalcea, Jia Deng
European Conference on Computer Vision (ECCV), 2016.
[paper][code]

We propose a structured matching of phrases and regions that encourages the semantic relations between phrases to agree with the visual relations between regions.

Line: Large-scale information network embedding
Jian Tang, Meng Qu, Mingzhe Wang, Ming Zhang, Jun Yan, Qiaozhu Mei
Proceedings of the 24th international conference on world wide web (WWW), 2015.
[paper][code]

We propose a novel network embedding method called the "LINE," which is suitable for arbitrary types of information networks: undirected, directed, and/or weighted.

Service

Reviewer: CVPR, ICCV, ECCV, NeurIPS, ICML

Template Credit: Jon Barron