Learning Data Representation by Large-Scale Neighbor Embedding

  • Sedov, Denis (Project Member)
  • Yang, Zhirong (Principal investigator)

Project Details


Machine learning has been increasingly influencing our life. The performance of machine learning methods heavily depends on the data representation. In this project we will develop a generic unsupervised method for learning data representation called Neighbor Embedding (NE). In recent practice we find that with large amount of neighborhood information, Neighbor Embedding significantly outperforms small-scale NE and many other existing approaches. In the development, we will emphasize scalable formulation and algorithms based on information divergences besides parallel computing tools. As a result, we can harness the fruitful gain from big data and thus our robust, efficient and convenient software will greatly facilitate data analysis in a wide range of research and applications.
Short titleYang Zhirong AT-kulut
Effective start/end date01/09/201731/12/2020


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.