Cross-Validation and Uncertainty Determination for Randomized Neural Networks with Applications to Mobile Sensors
This work addresses the challenge of improving prediction accuracy and quantifying uncertainty for randomized neural networks, which is important for researchers and practitioners deploying AI on resource-constrained mobile sensors.
This paper explores supervised learning with randomized neural networks, particularly for mobile sensors with limited resources and non-stationary data. It proposes a cross-validation approach to manage the inherent randomness and improve out-of-sample performance, along with a computationally efficient two-stage estimation method for determining confidence intervals for the prediction error.
Randomized artificial neural networks such as extreme learning machines provide an attractive and efficient method for supervised learning under limited computing ressources and green machine learning. This especially applies when equipping mobile devices (sensors) with weak artificial intelligence. Results are discussed about supervised learning with such networks and regression methods in terms of consistency and bounds for the generalization and prediction error. Especially, some recent results are reviewed addressing learning with data sampled by moving sensors leading to non-stationary and dependent samples. As randomized networks lead to random out-of-sample performance measures, we study a cross-validation approach to handle the randomness and make use of it to improve out-of-sample performance. Additionally, a computationally efficient approach to determine the resulting uncertainty in terms of a confidence interval for the mean out-of-sample prediction error is discussed based on two-stage estimation. The approach is applied to a prediction problem arising in vehicle integrated photovoltaics.