Semi-Tensor-Product Based Convolutional Neural Networks
By: Daizhan Cheng
Potential Business Impact:
Makes computers see better without fake data.
The semi-tensor product (STP) of vectors is a generalization of conventional inner product of vectors, which allows the factor vectors to of different dimensions. This paper proposes a domain-based convolutional product (CP). Combining domain-based CP with STP of vectors, a new CP is proposed. Since there is no zero or any other padding, it can avoid the junk information caused by padding. Using it, the STP-based convolutional neural network (CNN) is developed. Its application to image and third order signal identifications is considered.
Similar Papers
On Dimension-Free Transformer: An Application of STP to AI
Machine Learning (CS)
Makes computer learning work with any size information.
A Dimension-Keeping Semi-Tensor Product Framework for Compressed Sensing
Information Theory
Makes pictures clearer with less data.
Second-Order Tensorial Partial Differential Equations on Graphs
Machine Learning (CS)
Predicts traffic jams faster by understanding road networks.