DeepAI AI Chat
Log In Sign Up

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

by   Zhe-Cheng Fan, et al.

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.


page 8

page 10


Projections onto hyperbolas or bilinear constraint sets in Hilbert spaces

Sets of bilinear constraints are important in various machine learning m...

Pyramid Vector Quantization for Deep Learning

This paper explores the use of Pyramid Vector Quantization (PVQ) to redu...

Neural Similarity Learning

Inner product-based convolution has been the founding stone of convoluti...

An Order Preserving Bilinear Model for Person Detection in Multi-Modal Data

We propose a new order preserving bilinear framework that exploits low-r...

OMBA: User-Guided Product Representations for Online Market Basket Analysis

Market Basket Analysis (MBA) is a popular technique to identify associat...

Improved method for finding optimal formulae for bilinear maps in a finite field

In 2012, Barbulescu, Detrey, Estibals and Zimmermann proposed a new fram...