Towards Knowledge Uncertainty Estimation for Open Set Recognition

Abstract

Uncertainty is ubiquitous and happens in every single prediction of Machine Learning models. The ability to estimate and quantify the uncertainty of individual predictions is arguably relevant, all the more in safety-critical applications. Real-world recognition poses multiple challenges since a model’s knowledge about physical phenomenon is not complete, and observations are incomplete by definition. However, Machine Learning algorithms often assume that train and test data distributions are the same and that all testing classes are present during training. A more realistic scenario is the Open Set Recognition, where unknown classes can be submitted to an algorithm during testing. In this paper, we propose a Knowledge Uncertainty Estimation (KUE) method to quantify knowledge uncertainty and reject out-of-distribution inputs. Additionally, we quantify and distinguish aleatoric and epistemic uncertainty with the classical information-theoretical measures of entropy by means of ensemble techniques. We performed experiments on four datasets with different data modalities and compared our results with distance-based classifiers, SVM-based approaches and ensemble techniques using entropy measures. Overall, the effectiveness of KUE in distinguishing in- and out-distribution inputs obtained better results in most cases and was at least comparable in others. Furthermore, a classification with rejection option based on a proposed combination strategy between different measures of uncertainty is an application of uncertainty with proven results.

Uncertainty-Based Rejection in Machine Learning: Implications for Model Development and Interpretability

“…To define τ k , we used a 95% value of the training uncertainty values, meaning that τ k = P 95% [KUE]. A detailed description of this approach is available in [24].…”

“…Traditional methods, such as Kernel Density Estimation (KDE), can be used to estimate densities, and often, threshold-based methods are applied on top of the density where a classifier can refuse to predict a test input in that region [23]. In this context, Knowledge Uncertainty Estimation (KUE) [24] learns the feature density estimation from the training data, to reject test inputs that represent a density different from the training dataset. For a test input x i , represented by P-dimensional feature vectors, where f j ∈ { f 1 , .…”

Exploiting Epistemic Uncertainty at Inference Time for Early-Exit Power Saving

“…There are a number of methods, like this work, which focus on the inference process. KNN and distance metrics for anomaly detection have been studied previously [18,5,26], and recently have seen use in OOD and adversarial detection [35,1,39]. Furthermore, the notion of rejecting classification has been studied in depth [14,22,4].…”

Application of Machine Learning to Express Measurement Uncertainty

“…These concepts primarily deal with model uncertainty [22][23][24] (inverse propagation problem) over measurement uncertainty [25]. Recently, researchers have had increased interest in quantifying the uncertainty related to ML models [26][27][28]. In the systematic literature reviews of ML and deep learning applications in smart cities [11], and the COVID-19 epidemic [12], the accuracy parameter received the most attention from the reviewed articles (28.1% and 48.8%, respectively).…”