Uncertainty Propagation in Deep Neural Networks Using Extended Kalman Filtering
This work addresses uncertainty estimation in deep learning, which is crucial for safety-critical applications, but it is incremental as it builds on existing methods with computational improvements.
The paper tackled the problem of propagating and quantifying input uncertainty in deep neural networks by using Extended Kalman Filtering, achieving results comparable to existing methods while significantly reducing computational overhead and incorporating model error into output uncertainty.
Extended Kalman Filtering (EKF) can be used to propagate and quantify input uncertainty through a Deep Neural Network (DNN) assuming mild hypotheses on the input distribution. This methodology yields results comparable to existing methods of uncertainty propagation for DNNs while lowering the computational overhead considerably. Additionally, EKF allows model error to be naturally incorporated into the output uncertainty.