LGCVMLJul 14, 2022

Single Model Uncertainty Estimation via Stochastic Data Centering

arXiv:2207.07235v232 citationsh-index: 49Has Code
Originality Incremental advance
AI Analysis

This provides a more efficient way to estimate uncertainties for scientific and engineering applications, though it is incremental as it builds on existing ensemble and NTK concepts.

The paper tackles uncertainty estimation in deep neural networks by introducing a method that uses a single model to approximate ensemble behavior through input shifts, achieving superior performance on benchmarks like outlier rejection and calibration under distribution shift.

We are interested in estimating the uncertainties of deep neural networks, which play an important role in many scientific and engineering problems. In this paper, we present a striking new finding that an ensemble of neural networks with the same weight initialization, trained on datasets that are shifted by a constant bias gives rise to slightly inconsistent trained models, where the differences in predictions are a strong indicator of epistemic uncertainties. Using the neural tangent kernel (NTK), we demonstrate that this phenomena occurs in part because the NTK is not shift-invariant. Since this is achieved via a trivial input transformation, we show that this behavior can therefore be approximated by training a single neural network -- using a technique that we call $Δ-$UQ -- that estimates uncertainty around prediction by marginalizing out the effect of the biases during inference. We show that $Δ-$UQ's uncertainty estimates are superior to many of the current methods on a variety of benchmarks -- outlier rejection, calibration under distribution shift, and sequential design optimization of black box functions. Code for $Δ-$UQ can be accessed at https://github.com/LLNL/DeltaUQ

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes