Attentional Neural Network: Feature Selection Using Cognitive Feedback
This work addresses feature selection challenges in machine learning, particularly for high-noise scenarios, though it appears incremental as it builds on existing neural network and attention concepts.
The authors tackled the problem of feature selection in noisy or difficult segmentation tasks by integrating top-down cognitive bias with bottom-up feature extraction in a neural network framework, achieving competitive classification accuracy on the MNIST variation dataset and high success rates in disentangling overlaid digits.
Attentional Neural Network is a new framework that integrates top-down cognitive bias and bottom-up feature extraction in one coherent architecture. The top-down influence is especially effective when dealing with high noise or difficult segmentation problems. Our system is modular and extensible. It is also easy to train and cheap to run, and yet can accommodate complex behaviors. We obtain classification accuracy better than or competitive with state of art results on the MNIST variation dataset, and successfully disentangle overlaid digits with high success rates. We view such a general purpose framework as an essential foundation for a larger system emulating the cognitive abilities of the whole brain.