Evolution of active categorical image classification via saccadic eye movement
This work addresses computational efficiency in image classification for AI systems, but it is incremental as it builds on existing concepts like saccadic movements and evolutionary computation.
The paper tackles the challenge of high computational cost in image classification by proposing an active categorical classifier (ACC) that selectively scans portions of images, inspired by saccadic eye movements. It demonstrates the ACC's ability to classify MNIST digits after viewing only a fraction of pixels, providing a proof-of-concept on noisy multi-class data.
Pattern recognition and classification is a central concern for modern information processing systems. In particular, one key challenge to image and video classification has been that the computational cost of image processing scales linearly with the number of pixels in the image or video. Here we present an intelligent machine (the "active categorical classifier," or ACC) that is inspired by the saccadic movements of the eye, and is capable of classifying images by selectively scanning only a portion of the image. We harness evolutionary computation to optimize the ACC on the MNIST hand-written digit classification task, and provide a proof-of-concept that the ACC works on noisy multi-class data. We further analyze the ACC and demonstrate its ability to classify images after viewing only a fraction of the pixels, and provide insight on future research paths to further improve upon the ACC presented here.