Single Model Uncertainty Estimation via Stochastic Data Centering
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 $\Delta-$UQ -- that estimates uncertainty around prediction by marginalizing out the effect of the biases during inference. We show that $\Delta-$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 $\Delta-$UQ can be accessed at https://github.com/LLNL/DeltaUQ
Code (1)
Tasks
modelMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Accurate and Scalable Estimation of Epistemic Uncertainty for Graph Neural Networks
Safe deployment of graph neural networks (GNNs) under distribution shift requires models to provide accurate confidence indicators (CI). However, while it is well-known in computer vision that CI quality diminishes under…
Uncertainty QuantificationAccurate and Scalable Estimation of Epistemic Uncertainty for Graph Neural Networks
While graph neural networks (GNNs) are widely used for node and graph representation learning tasks, the reliability of GNN uncertainty estimates under distribution shifts remains relatively under-explored. Indeed, while…
Graph ClassificationGraph Representation LearningOut-of-Distribution DetectionRepresentation LearningStochastic Actor-Critic: Mitigating Overestimation via Temporal Aleatoric Uncertainty
Off-policy actor-critic methods in reinforcement learning train a critic with temporal-difference updates and use it as a learning signal for the policy (actor). This design typically achieves higher sample efficiency th…
Computational EfficiencyReinforcement LearningStochastic Order Learning: An Approach to Rank Estimation Using Noisy Data
Rank estimation under label noise poses a fundamental challenge, as ordinal annotations often exhibit structured uncertainty rather than simple label corruption. In this paper, we reformulate rank estimation with noisy o…
Interacting Contour Stochastic Gradient Langevin Dynamics
We propose an interacting contour stochastic gradient Langevin dynamics (ICSGLD) sampler, an embarrassingly parallel multiple-chain contour stochastic gradient Langevin dynamics (CSGLD) sampler with efficient interaction…