HyperMix: Out-of-Distribution Detection and Classification in Few-Shot Settings
This addresses the challenge of OOD detection for real-world ML systems in data-scarce scenarios, representing an incremental improvement over existing methods.
The paper tackled the problem of out-of-distribution detection in few-shot settings, where models have limited in-distribution samples, and demonstrated that HyperMix, a hypernetwork framework with Mixup and outlier exposure, significantly outperformed other methods on datasets like CIFAR-FS and MiniImageNet.
Out-of-distribution (OOD) detection is an important topic for real-world machine learning systems, but settings with limited in-distribution samples have been underexplored. Such few-shot OOD settings are challenging, as models have scarce opportunities to learn the data distribution before being tasked with identifying OOD samples. Indeed, we demonstrate that recent state-of-the-art OOD methods fail to outperform simple baselines in the few-shot setting. We thus propose a hypernetwork framework called HyperMix, using Mixup on the generated classifier parameters, as well as a natural out-of-episode outlier exposure technique that does not require an additional outlier dataset. We conduct experiments on CIFAR-FS and MiniImageNet, significantly outperforming other OOD methods in the few-shot regime.