Single-Model Uncertainties for Deep Learning
This provides a computationally efficient method for uncertainty estimation in deep learning, which is crucial for applications like safety-critical systems, but it is incremental as it builds on existing uncertainty quantification techniques.
The paper tackled the problem of estimating both aleatoric and epistemic uncertainty in deep neural networks without requiring ensembles or retraining, achieving competitive performance with well-calibrated prediction intervals.
We provide single-model estimates of aleatoric and epistemic uncertainty for deep neural networks. To estimate aleatoric uncertainty, we propose Simultaneous Quantile Regression (SQR), a loss function to learn all the conditional quantiles of a given target variable. These quantiles can be used to compute well-calibrated prediction intervals. To estimate epistemic uncertainty, we propose Orthonormal Certificates (OCs), a collection of diverse non-constant functions that map all training samples to zero. These certificates map out-of-distribution examples to non-zero values, signaling epistemic uncertainty. Our uncertainty estimators are computationally attractive, as they do not require ensembling or retraining deep models, and achieve competitive performance.