Gaussian Latent Representations for Uncertainty Estimation using Mahalanobis Distance in Deep Classifiers
This addresses uncertainty estimation for deep classifiers, particularly in computer vision applications like microorganism classification, but is incremental as it builds on existing Mahalanobis distance methods.
The paper tackled the problem of estimating classification uncertainty and detecting out-of-distribution samples by introducing a lightweight regularization method for Mahalanobis distance-based prediction, achieving state-of-the-art results on OOD detection benchmarks with minimal inference time.
Recent works show that the data distribution in a network's latent space is useful for estimating classification uncertainty and detecting Out-of-distribution (OOD) samples. To obtain a well-regularized latent space that is conducive for uncertainty estimation, existing methods bring in significant changes to model architectures and training procedures. In this paper, we present a lightweight, fast, and high-performance regularization method for Mahalanobis distance-based uncertainty prediction, and that requires minimal changes to the network's architecture. To derive Gaussian latent representation favourable for Mahalanobis Distance calculation, we introduce a self-supervised representation learning method that separates in-class representations into multiple Gaussians. Classes with non-Gaussian representations are automatically identified and dynamically clustered into multiple new classes that are approximately Gaussian. Evaluation on standard OOD benchmarks shows that our method achieves state-of-the-art results on OOD detection with minimal inference time, and is very competitive on predictive probability calibration. Finally, we show the applicability of our method to a real-life computer vision use case on microorganism classification.