Abstract: This thesis comes up with a fast method to train convolutional neural networks (CNNs) using recursive least squares (RLS) algorithm in conjunction with the back-propagation learning. The recursive updating equations for CNNs are derived via the back-propagation method and normal equations. This method does not need the choice of a learning rate nor suffer from speed-accuracy trade-off. Additionally, it is much faster than the conventional gradient-based methods in a sense that it needs less epochs to converge. The performance of the proposed method together with those of the standard gradient-based methods are compared on the MNIST handwritten digits and Fashion-MNIST clothes databases. The simulation results show that the proposed RLS-based training method requires only one epoch to meet the error goal during the training phase while offering comparable accuracy on the testing data sets.

Adviser: Mahmood Azimi-Sadjadi
Co-Adviser: N/A
Non-ECE Member: Iuliana Oprea, Mathematics
Member 3: Ali Pezeshki, ECE
Addional Members: N/A

Publications:
N/A

Program of Study:
ECE455
ECE512
ECE520
ECE656
ECE513
ECE514
ECE699
MATH560