Revisiting Inlier and Outlier Specification for Improved Out-of-Distribution Detection

@article{Narayanaswamy2022RevisitingIA,
  title={Revisiting Inlier and Outlier Specification for Improved Out-of-Distribution Detection},
  author={Vivek Sivaraman Narayanaswamy and Yamen Mubarka and Rushil Anirudh and Deepta Rajan and Andreas Spanias and Jayaraman J. Thiagarajan},
  journal={ArXiv},
  year={2022},
  volume={abs/2207.05286}
}
Accurately detecting out-of-distribution (OOD) data with varying levels of semantic and covariate shifts with respect to the in-distribution (ID) data is critical for deployment of safe and reliable models. This is particularly the case when dealing with highly conse-quential applications (e.g. medical imaging, self-driving cars, etc). The goal is to design a detector that can accept meaningful variations of the ID data, while also rejecting examples from OOD regimes. In practice, this dual… 

References

SHOWING 1-10 OF 30 REFERENCES

Fine-grained Out-of-Distribution Detection with Mixup Outlier Exposure

A new DNN training algorithm is proposed, Mixup Outlier Exposure (MixupOE), which leverages an outlier distribution and principles from vicinal risk minimization and can consistently improve fine-grained detection performance, establishing a strong baseline in these more realistic and challenging OOD detection settings.

ATOM: Robustifying Out-of-Distribution Detection Using Outlier Mining

It is shown that, by mining informative auxiliary OOD data, one can significantly improve OOD detection performance, and somewhat surprisingly, generalize to unseen adversarial attacks.

VOS: Learning What You Don't Know by Virtual Outlier Synthesis

VOS is presented, a novel framework for OOD detection by adaptively synthesizing virtual outliers that can meaningfully regularize the model’s decision boundary during training, which achieves stateof-the-art performance on both object detection and image classification models.

Detecting Out-of-Distribution Examples with In-distribution Examples and Gram Matrices

It is found that characterizing activity patterns by Gram matrices and identifying anomalies in gram matrix values can yield high OOD detection rates, and this method generally performs better than or equal to state-of-the-art Ood detection methods.

Generalized ODIN: Detecting Out-of-Distribution Image Without Learning From Out-of-Distribution Data

This work bases its work on a popular method ODIN, proposing two strategies for freeing it from the needs of tuning with OoD data, while improving its OoD detection performance, and proposing to decompose confidence scoring as well as a modified input pre-processing method.

OoD-Bench: Benchmarking and Understanding Out-of-Distribution Generalization Datasets and Algorithms

This work identifies and measure two distinct kinds of distribution shifts that are ubiquitous in various datasets and compares OoD generalization algorithms across two groups of benchmarks, revealing their strengths on one distribution shift as well as limitations on the other distribution shift.

Deep Anomaly Detection with Outlier Exposure

In extensive experiments on natural language processing and small- and large-scale vision tasks, it is found that Outlier Exposure significantly improves detection performance and that cutting-edge generative models trained on CIFar-10 may assign higher likelihoods to SVHN images than to CIFAR-10 images; OE is used to mitigate this issue.

A Simple Fix to Mahalanobis Distance for Improving Near-OOD Detection

On a wide selection of challenging vision, language, and biology OOD benchmarks, it is shown that RMD meaningfully improves upon MD performance (by up to 15% AUROC on genomics OOD).

A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks

This paper proposes a simple yet effective method for detecting any abnormal samples, which is applicable to any pre-trained softmax neural classifier, and obtains the class conditional Gaussian distributions with respect to (low- and upper-level) features of the deep models under Gaussian discriminant analysis.

Energy-based Out-of-distribution Detection

This work proposes a unified framework for OOD detection that uses an energy score, and shows that energy scores better distinguish in- and out-of-distribution samples than the traditional approach using the softmax scores.