Backpropagation with N-D Vector-Valued Neurons Using Arbitrary Bilinear Products

Abstract

Vector-valued neural learning has emerged as a promising direction in deep learning recently. Traditionally, training data for neural networks (NNs) are formulated as a vector of scalars; however, its performance may not be optimal since associations among adjacent scalars are not modeled. In this paper, we propose a new vector neural architecture called the Arbitrary BIlinear Product Neural Network (ABIPNN), which processes information as vectors in each neuron, and the feedforward projections are defined using arbitrary bilinear products. Such bilinear products can include circular convolution, seven-dimensional vector product, skew circular convolution, reversed- time circular convolution, or other new products not seen in previous work. As a proof-of-concept, we apply our proposed network to multispectral image denoising and singing voice sepa- ration. Experimental results show that ABIPNN gains substantial improvements when compared to conventional NNs, suggesting that associations are learned during training.

arXiv comment14 pages, 8 figures, 3 tables

References in corpus (7)

• Adam: A Method for Stochastic Optimization
• Joint Optimization of Masks and Deep Recurrent Neural Networks for Monaural Source Separation
• Deep Complex Networks
• On Variance Reduction in Stochastic Gradient Descent and its Asynchronous Variants
• Complex Numbers in n Dimensions
• Polar $n$-Complex and $n$-Bicomplex Singular Value Decomposition and Principal Component Pursuit
• On Systems of Linear Quaternion Functions