Supervised Feature Selection with Neuron Evolution in Sparse Neural Networks
Preprint
Abstract
This paper proposes a novel supervised feature selection method named NeuroFS. NeuroFS introduces dynamic neuron evolution in the training process of a sparse neural network to find an informative set of features. By evaluating NeuroFS on real-world benchmark datasets, we demonstrated that it achieves the highest ranking-based score among the considered state-of-the-art supervised feature selection models. However, due to the general lack of knowledge on optimally implementing sparse neural networks during training, NeuroFS does not take full advantage of its theoretical high computational and memory advantages. We let the development of this challenging research direction for future work, hopefully, in a greater joint effort of the community.Keywords
All Related Versions
This publication has 16 references indexed in Scilit:
- FsNet: Feature Selection Network on High-dimensional Biological Data2023
- A brain-inspired algorithm for training highly sparse neural networksMachine Learning, 2022
- Deep unsupervised feature selection by discarding nuisance and correlated featuresNeural Networks, 2022
- Quick and robust feature selection: the strength of energy-efficient sparse training for autoencodersMachine Learning, 2021
- Feature selection with multi-view data: A surveyInformation Fusion, 2019
- NeST: A Neural Network Synthesis Tool Based on a Grow-and-Prune ParadigmIEEE Transactions on Computers, 2019
- Gene Selection for Cancer Classification using Support Vector MachinesMachine Learning, 2002
- Wrappers for feature subset selectionArtificial Intelligence, 1997
- Neural-network feature selectorIEEE Transactions on Neural Networks, 1997
- Using mutual information for selecting features in supervised neural net learningIEEE Transactions on Neural Networks, 1994