Multi-kernel Classification Machine with Universum learning

Universum learning can reflect priori knowledge about application domain and improve classification performances. Since multi-kernel classification machine with reduced complexity, i.e., NMKMHKS, has a high performance to reduce both the time and space complexities of multiple kernel learning (MKL), so we introduce Universum learning into NMKMHKS and propose a Universum-based NMKMHKS (Uni-NMKMHKS).

Changming Zhu received the B.S degree and PHD degree in Department of Computer Science and Engineering, East China University of Science and Technology (ECUST), China, 2010 and 2015 respectively. Currently he is a teacher in College of Information Engineering, Shanghai Maritime University. His research interests focus on neural computing.