CNNtention: Can CNNs do better with Attention?
This is an incremental analysis for the deep learning community, guiding model selection based on application needs.
The paper compares traditional CNNs with attention-augmented CNNs on an image classification task, evaluating their performance, accuracy, and computational efficiency to highlight trade-offs between localized feature extraction and global context capture.
Convolutional Neural Networks (CNNs) have been the standard for image classification tasks for a long time, but more recently attention-based mechanisms have gained traction. This project aims to compare traditional CNNs with attention-augmented CNNs across an image classification task. By evaluating and comparing their performance, accuracy and computational efficiency, the project will highlight benefits and trade-off of the localized feature extraction of traditional CNNs and the global context capture in attention-augmented CNNs. By doing this, we can reveal further insights into their respective strengths and weaknesses, guide the selection of models based on specific application needs and ultimately, enhance understanding of these architectures in the deep learning community. This was our final project for CS7643 Deep Learning course at Georgia Tech.