Timezone: »

Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awareness
Jeremiah Liu · Zi Lin · Shreyas Padhy · Dustin Tran · Tania Bedrax Weiss · Balaji Lakshminarayanan

Thu Dec 10 09:00 AM -- 11:00 AM (PST) @ Poster Session 5 #1615

Bayesian neural networks (BNN) and deep ensembles are principled approaches to estimate the predictive uncertainty of a deep learning model. However their practicality in real-time, industrial-scale applications are limited due to their heavy memory and inference cost. This motivates us to study principled approaches to high-quality uncertainty estimation that require only a single deep neural network (DNN). By formalizing the uncertainty quantification as a minimax learning problem, we first identify input distance awareness, i.e., the model’s ability to quantify the distance of a testing example from the training data in the input space, as a necessary condition for a DNN to achieve high-quality (i.e., minimax optimal) uncertainty estimation. We then propose Spectral-normalized Neural Gaussian Process (SNGP), a simple method that improves the distance-awareness ability of modern DNNs, by adding a weight normalization step during training and replacing the output layer. On a suite of vision and language understanding tasks and on modern architectures (Wide-ResNet and BERT), SNGP is competitive with deep ensembles in prediction, calibration and out-of-domain detection, and outperforms the other single-model approaches.

Author Information

Jeremiah Liu (Google Research / Harvard)
Zi Lin (Google)
Shreyas Padhy (Google)
Dustin Tran (Google Brain)
Tania Bedrax Weiss (Google)
Balaji Lakshminarayanan (Google Brain)

More from the Same Authors