Conditional Gumbel-Softmax for constrained feature selection with application to node selection in wireless sensor networks
This addresses constrained node selection in wireless sensor networks, particularly for wearable brain-computer interfaces, with incremental improvements over existing methods.
The paper tackled the problem of selecting optimal feature subsets under pairwise constraints, introducing Conditional Gumbel-Softmax for end-to-end learning, and validated it on a wireless EEG sensor network for a motor execution task, showing performance variations with constraint stringency and comparison to a greedy method.
In this paper, we introduce Conditional Gumbel-Softmax as a method to perform end-to-end learning of the optimal feature subset for a given task and deep neural network (DNN) model, while adhering to certain pairwise constraints between the features. We do this by conditioning the selection of each feature in the subset on another feature. We demonstrate how this approach can be used to select the task-optimal nodes composing a wireless sensor network (WSN) while ensuring that none of the nodes that require communication between one another have too large of a distance between them, limiting the required power spent on this communication. We validate this approach on an emulated Wireless Electroencephalography (EEG) Sensor Network (WESN) solving a motor execution task. We analyze how the performance of the WESN varies as the constraints are made more stringent and how well the Conditional Gumbel-Softmax performs in comparison with a heuristic, greedy selection method. While the application focus of this paper is on wearable brain-computer interfaces, the proposed methodology is generic and can readily be applied to node deployment in wireless sensor networks and constrained feature selection in other applications as well.