Sketching Curvature for Efficient Out-of-Distribution Detection for Deep Neural Networks
This addresses the need for efficient and accurate safeguards in real-time decision-making systems, though it is incremental as it builds on prior curvature-based methods.
The paper tackles the problem of detecting out-of-distribution inputs for deep neural networks by proposing SCOD, a framework that uses matrix sketching to approximate curvature for efficient uncertainty estimation, achieving comparable or better performance with lower computational cost than existing methods.
In order to safely deploy Deep Neural Networks (DNNs) within the perception pipelines of real-time decision making systems, there is a need for safeguards that can detect out-of-training-distribution (OoD) inputs both efficiently and accurately. Building on recent work leveraging the local curvature of DNNs to reason about epistemic uncertainty, we propose Sketching Curvature of OoD Detection (SCOD), an architecture-agnostic framework for equipping any trained DNN with a task-relevant epistemic uncertainty estimate. Offline, given a trained model and its training data, SCOD employs tools from matrix sketching to tractably compute a low-rank approximation of the Fisher information matrix, which characterizes which directions in the weight space are most influential on the predictions over the training data. Online, we estimate uncertainty by measuring how much perturbations orthogonal to these directions can alter predictions at a new test input. We apply SCOD to pre-trained networks of varying architectures on several tasks, ranging from regression to classification. We demonstrate that SCOD achieves comparable or better OoD detection performance with lower computational burden relative to existing baselines.