Backup

Yang Yu (Y. Yu) Can be pronounced as "young you" Ph.D., Assistant Researcher LAMDA Group, Department of Computer Science, Nanjing University Office: 919, Computer Science Building, Xianlin Campus email: yuy@nju.edu.cn (alternatives: yuy@lamda.nju.edu.cn, eyounx@gmail.com)

Research

• ensemble learning (my related publications)
• theoretical foundation of evolutionary algorithms (my related publications)
• learning for robotics (my related publications)

• General approaches to running time analysis of metaheuristic optimization algorithms: (with Chao Qian and Zhi-Hua Zhou)
  The running time, or the computational complexity, of evolutionary algorithms and other metaheuristic algorithms is one of the most important theoretical issue to understand these algorithms.
  • We developed the convergence-based approach (PDF-aij08) that can be applied to obtain running time bounds of a large range of metaheuristic algorithms.
  • For sophisticated metaheuristic algorithms, it is usually hard to have a fine analysis. Thus we developed the switch analysis (tr11|PDF-ppsn10) that can lead to much tighter running time bounds.

• Approximation ability of evolutionary optimization: (with Xin Yao and Zhi-Hua Zhou)
  Evolutionary algorithms are most commonly used to obtain good-enough solutions rather than optimal solutions in practice, which relates to their approximation ability. We developed the SEIP framework (PDF-aij12) to characterize the approximation ability of evolutionary algorithms, and showed that they can achieve the currently best-achievable approximation ratio for the k-set cover problem, which also reveals the advantage of evolutionary algorithms over the well known greedy algorithm.

• The role of diversity in ensemble learning (with Yu-Feng Li and Zhi-Hua Zhou)
  Ensemble learning is a machine learning paradigm that achieves the state-of-the-art performance. Diversity was believed to be a key to a good performance of an ensemble approach, which, however, only provides heuristic ideas for the existing ensemble approaches. By the diversity regularized machine (PDF-ijcai11), we show that diversity plays a role of regularization as in popular statistical learning approaches.

Prototypes

• Watering robot: a robot that explores an unknown environment to water plants.

Recent Professional Activities

• Invited Speaker: ICONIP'2011 Workshop on Recent Advances in Nature Inspired Computation and Its Applications, Nov. 2011, Hangzhou, China
• Invited Speaker: IWNICA'11 (The 2011 International Workshop on Nature Inspired Computation and Applications, April 2011, Hefei, China):

Teaching Asistant

• Introduction to Data Mining. (for undergraduate students. Spring, 2012) >>>Teaching Assistant Page>>>
  
• Data mining. (for graduate students. Fall, 2007) Teaching Assistant Page
  
• Algorithm Design and Analysis. (for undergraduate students. Fall, 2005)