Dr. Wang's paper, "Model Accuracy and Runtime Tradeoff in Distributed Deep Learning: A Systematic Study," will receive runner-up for best research paper at the IEEE International Conference on Data Mining (ICDM) 2016, which will be held in Barcelona, Spain from Dec. 12 to Dec. 15. This year, the acceptance rate for regular papers is just 8.5%
The paper is in collaboration with the researchers in IBM T.J. Watson Research, where they presented Rudra, a parameter server-based distributed computing framework tuned for training large-scale deep neural networks. They also did a systematic study on the model accuracy and runtime tradeoff with Rudra. Deep learning is also a strategy that could be very helpful for health informatics research problems such as predictive modeling. Dr. Wang's study will be crucial to understanding the behavior of deep learning and how to apply them in a correct way.