Browsing by Author "Bao, Yi-Ting"

Now showing 1 - 2 of 2

TENSOR CLASSIFICATION NETWORK
(2015)
Bao, Yi-Ting
;
Chien, Jen-Tzung
;
電機學院
;
College of Electrical and Computer Engineering
The growing interests in multi-way data analysis have made the tensor factorization and classification a crucial issue in machine learning for signal processing. Conventional neural network (NN) classifier is estimated from a set of input vectors. The multi-way data are unfolded as high-dimensional vectors for model training. The classification performance is constrained because the neighboring temporal or spatial information in different ways is lost in the trained NN classifier. More parameters are required to learn the complicated data structure. This paper presents a new tensor classification network (TCN) which combines the tensor factorization and the NN classification for multi-way feature extraction and classification. We generalize from NN classifier to TCN where the multi-way information is preserved through projecting the input tensors into latent tensors. The projection over tensor space is efficiently characterized so that a very compact classifier can be achieved. Experimental results on image recognition demonstrate that TCN attains comparable classification performance but with very few parameters (as small as 6.7%) compared with NN classifier.
Tensor-Factorized Neural Networks
(2018-05-01)
Chien, Jen-Tzung
;
Bao, Yi-Ting
;
電機工程學系
;
Department of Electrical and Computer Engineering
The growing interests in multiway data analysis and deep learning have drawn tensor factorization (TF) and neural network (NN) as the crucial topics. Conventionally, the NN model is estimated from a set of one-way observations. Such a vectorized NN is not generalized for learning the representation from multiway observations. The classification performance using vectorized NN is constrained, because the temporal or spatial information in neighboring ways is disregarded. More parameters are required to learn the complicated data structure. This paper presents a new tensor-factorized NN (TFNN), which tightly integrates TF and NN for multiway feature extraction and classification under a unified discriminative objective. This TFNN is seen as a generalized NN, where the affine transformation in an NN is replaced by the multilinear and multiway factorization for tensor-based NN. The multiway information is preserved through layerwise factorization. Tucker decomposition and nonlinear activation are performed in each hidden layer. The tensor-factorized error backpropagation is developed to train TFNN with the limited parameter size and computation time. This TFNN can be further extended to realize the convolutional TFNN (CTFNN) by looking at small subtensors through the factorized convolution. Experiments on real-world classification tasks demonstrate that TFNN and CTFNN attain substantial improvement when compared with an NN and a convolutional NN, respectively.