DartsReNet: Exploring new RNN cells in ReNet architectures

• Published in: arXiv.org
• Main Authors: Moser, Brian, Raue, Federico, Hees, Jörn, Dengel, Andreas
• Format: Paper Journal Article
• Language: English
• Published: Ithaca Cornell University Library, arXiv.org 11.04.2023
• Subjects: Computer Science - Artificial Intelligence; Computer Science - Computer Vision and Pattern Recognition; Computer Science - Learning; Image classification; Recurrent neural networks
• ISSN: 2331-8422
• DOI: 10.48550/arxiv.2304.05838

We present new Recurrent Neural Network (RNN) cells for image classification using a Neural Architecture Search (NAS) approach called DARTS. We are interested in the ReNet architecture, which is a RNN based approach presented as an alternative for convolutional and pooling steps. ReNet can be defined using any standard RNN cells, such as LSTM and GRU. One limitation is that standard RNN cells were designed for one dimensional sequential data and not for two dimensions like it is the case for image classification. We overcome this limitation by using DARTS to find new cell designs. We compare our results with ReNet that uses GRU and LSTM cells. Our found cells outperform the standard RNN cells on CIFAR-10 and SVHN. The improvements on SVHN indicate generalizability, as we derived the RNN cell designs from CIFAR-10 without performing a new cell search for SVHN.