# DropIn: Making Reservoir Computing Neural Networks Robust to Missing Inputs by Dropout

## TL;DR

DropIn introduces a training method for Reservoir Computing neural networks that enhances robustness to missing input features, maintaining high performance even with significant input loss.

## Contribution

The paper proposes DropIn, a novel dropout-based training approach that creates subnetworks capable of handling missing inputs in Reservoir Computing models.

## Key findings

- Maintains predictive performance with 20-50% missing inputs
- Effective in wireless sensor networks and ambient intelligence applications
- Outperforms traditional models in robustness to input loss

## Abstract

The paper presents a novel, principled approach to train recurrent neural networks from the Reservoir Computing family that are robust to missing part of the input features at prediction time. By building on the ensembling properties of Dropout regularization, we propose a methodology, named DropIn, which efficiently trains a neural model as a committee machine of subnetworks, each capable of predicting with a subset of the original input features. We discuss the application of the DropIn methodology in the context of Reservoir Computing models and targeting applications characterized by input sources that are unreliable or prone to be disconnected, such as in pervasive wireless sensor networks and ambient intelligence. We provide an experimental assessment using real-world data from such application domains, showing how the Dropin methodology allows to maintain predictive performances comparable to those of a model without missing features, even when 20\%-50\% of the inputs are not available.