Projects

Deep Learning & Machine Learning

How can we learn from massive amount of data? We work on designing models, algorithms, and systems for deep learning and machine learning. The topics include belief propagation, deep network embedding, data clustering, and regression.

Models, Algorithms and Systems:

  • Belief propagation: Belief Propagation (BP) is an efficient algorithm for probabilistic inference in graphical model. BP is used for anomaly detection based on the idea of "guilt by association": if we know that nodes of type "A" (say, males) tend to interact/date nodes of type "B" (females), we can infer the unknown gender of a node, by checking the gender of the majority of its contacts. We develop scalable BP algorithms, and models for complex interactions among many variables.
  • Deep learning: We develop models, applications, and algorithms for deep learning. Notable projects include deep network embedding, and lightweight deep learning.
  • Question answering: We develop question answering methods using big data and deep neural networks.
  • Data clustering: Given a matrix (or tensor), we develop scalable co-clustering methods which aims to simultaneously clusters both rows and columns of it.
  • Regression: We develop scalable algorithms for regression problems including logistic and linear regression.

Publication

  • Namyong Park, Eunjeong Kang, Minsu Park, Hajeong Lee, Hee-Gyung Kang, Hyung-Jin Yoon, and U Kang, "Predicting acute kidney injury in cancer patients using heterogeneous and irregular data", PLOS ONE, 2018. (to appear)
  • Junghwan Kim, Haekyu Park, Ji-Eun Lee, and U Kang, "SIDE: Representation Learning in Signed Directed Networks", The Web Conference (WWW) 2018, Lyon, France. [PDF] [BIBTEX]
  • Saehan Jo, Jaemin Yoo, and U Kang, "Fast and Scalable Distributed Loopy Belief Propagation on Real-World Graphs", 11th ACM International Conference on Web Search and Data Mining (WSDM) 2018, Los Angeles, CA, USA. [BIBTEX] [HOMEPAGE (CODE, DATA)] [PDF]
  • Jaemin Yoo, Saehan Jo, and U Kang, "Supervised Belief Propagation: Scalable Supervised Inference on Attributed Networks", IEEE International Conference on Data Mining (ICDM) 2017, New Orleans, USA. [BIBTEX] [HOMEPAGE (CODE, DATA)] [PDF]
  • Kyung-Min Kim, Jinhong Jung, Jihee Ryu, Ha-Myung Park, Joseph P.Joohee, Seokwoo Jeong, U Kang, and Sung-Hyon Myaeng, "A New Question Answering Approach with Conceptual Graphs", Conférence en Recherche d’Information et Applications (CORIA) 2017, Marseille, France. [BIBTEX] [PDF]
  • Min-Hee Jang, Christos Faloutsos, Sang-Wook Kim, U Kang, and Jiwoon Ha, "PIN-TRUST: Fast Trust Propagation Exploiting Positive, Implicit, and Negative Information", ACM International Conference on Information and Knowledge Management (CIKM) 2016, Indianapolis, Indiana, USA. [BIBTEX] [PDF]
  • Won-Jo Lee, Chae-Gyun Lim, U Kang, and Ho-Jin Choi, "An Extension of the Automatic Cross-Association Method with a 3-dimensional Matrix", Second International Conference on Big Data and Smart Computing (BigComp) 2015, Jeju, korea. [BIBTEX] [PDF]
  • Dongyeop Kang, Woosang Lim, Kijung Shin, Lee Sael, and U Kang, "Data/Feature Distributed Stochastic Coordinate Descent for Logistic Regression", 23rd ACM International Conference on Information and Knowledge Management (CIKM) 2014, Shaghai, China [BIBTEX] [PDF] [SUPPLEMENTARY DOCUMENT]
  • Danai Koutra, Tai-You Ke, U Kang, Duen Horng (Polo) Chau, Hsing-Kuo Kenneth Pao, and Christos Faloutsos, "Unifying Guilt-by-Association Approaches: Theorems and Fast Algorithms", European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD) 2011, Athens, Greece. (acceptance rate 20.2 %) [BIBTEX] [PDF]
  • Robson L. F. Cordeiro, Caetano Traina Jr., Agma J. M. Traina, Julio Lopez, U Kang, and Christos Faloutsos, "Clustering Very Large Multi-dimensional Datasets with MapReduce", ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) 2011, San Diego, CA, USA. (acceptance rate 17.5 %) [BIBTEX] [PDF]
  • U Kang, Duen Horng Chau, and Christos Faloutsos, "Mining Large Graphs: Algorithms, Inference, and Discoveries", IEEE International Conference on Data Engineering (ICDE) 2011, Hannover, Germany. (acceptance rate 19.8 %) [BIBTEX] [PDF]