Exploiting layerwise convexity of rectifier networks with sign constrained weights
MetadataShow full item record
© 2018 Elsevier Ltd By introducing sign constraints on the weights, this paper proposes sign constrained rectifier networks (SCRNs), whose training can be solved efficiently by the well known majorization–minimization (MM) algorithms. We prove that the proposed two-hidden-layer SCRNs, which exhibit negative weights in the second hidden layer and negative weights in the output layer, are capable of separating any number of disjoint pattern sets. Furthermore, the proposed two-hidden-layer SCRNs can decompose the patterns of each class into several clusters so that each cluster is convexly separable from all the patterns from the other classes. This provides a means to learn the pattern structures and analyse the discriminant factors between different classes of patterns. Experimental results are provided to show the benefits of sign constraints in improving classification performance and the efficiency of the proposed MM algorithm.
Showing items related by title, author, creator and subject.
An, Senjian; Ke, Q.; Bennamoun, M.; Boussaid, F.; Sohel, F. (2015)© Springer International Publishing Switzerland 2015. In this paper we introduce sign constrained rectifier networks (SCRN), demonstrate their universal classification power and illustrate their applications to pattern ...
Lim, King Hann; Seng, Kah Phooi; Ang, Li-Minn (2012)Lyapunov theory-based radial basis function neural network (RBFNN) is developed for traffic sign recognition in this paper to perform multiple inputs multiple outputs (MIMO) classification. Multidimensional input is ...
Hong, D.; Liu, Wan-Quan; Wu, X.; Pan, Z.; Su, J. (2016)Palmprint is usually captured with a touchless device. Due to the changes of angle and position of the palm in a capturing process, as well as the defocus of device, it is inevitable to have some distortions in translation, ...