Meta Learning Low Rank Covariance Factors for Energy-Based Deterministic Uncertainty
This work addresses uncertainty calibration for machine learning models in multi-task settings, representing an incremental improvement over existing methods.
The paper tackles the problem of sub-optimal covariance matrices in deterministic uncertainty estimation and out-of-distribution detection when dealing with a distribution over tasks, by proposing to meta-learn low-rank covariance factors using an attentive set encoder, resulting in well-calibrated predictive distributions under dataset shift.
Numerous recent works utilize bi-Lipschitz regularization of neural network layers to preserve relative distances between data instances in the feature spaces of each layer. This distance sensitivity with respect to the data aids in tasks such as uncertainty calibration and out-of-distribution (OOD) detection. In previous works, features extracted with a distance sensitive model are used to construct feature covariance matrices which are used in deterministic uncertainty estimation or OOD detection. However, in cases where there is a distribution over tasks, these methods result in covariances which are sub-optimal, as they may not leverage all of the meta information which can be shared among tasks. With the use of an attentive set encoder, we propose to meta learn either diagonal or diagonal plus low-rank factors to efficiently construct task specific covariance matrices. Additionally, we propose an inference procedure which utilizes scaled energy to achieve a final predictive distribution which is well calibrated under a distributional dataset shift.