ood detection method
Rethinking the Evaluation of Out-of-Distribution Detection: A Sorites Paradox
However, some marginal OOD samples actually have close semantic contents to the in-distribution (ID) sample, which makes determining the OOD sample a Sorites Paradox.
- Law (0.67)
- Information Technology (0.67)
- Information Technology > Artificial Intelligence > Natural Language (1.00)
- Information Technology > Artificial Intelligence > Vision (0.93)
- Information Technology > Sensing and Signal Processing > Image Processing (0.69)
- Information Technology > Artificial Intelligence > Machine Learning > Neural Networks > Deep Learning (0.67)
SeT AR: Out-of-Distribution Detection with Selective Low-Rank Approximation
SeT AR enhances OOD detection via post-hoc modification of the model's
- Asia > China > Shanghai > Shanghai (0.04)
- North America > United States (0.04)
- North America > Dominican Republic (0.04)
- Asia > Middle East > Israel (0.04)
- Research Report > New Finding (0.93)
- Research Report > Experimental Study (0.93)
The Best of Both Worlds: On the Dilemma of Out-of-distribution Detection
Work done during an internship at Tencent AI Lab.
- Europe > Spain > Andalusia > Granada Province > Granada (0.04)
- Asia > China > Tianjin Province > Tianjin (0.04)
A TT A: Anomaly-aware Test-Time Adaptation for Out-of-Distribution Detection in Segmentation
We quantify the drop in PEBAL's performance with the added domain shift and compare it to the performance when combined with our method or existing test-time adaptation methods such as
- Asia > China > Shanghai > Shanghai (0.04)
- Asia > Middle East > Israel (0.04)
- Europe > Netherlands > South Holland > Delft (0.04)
- Asia > Middle East > Jordan (0.04)
Appendix: BoostingOut-of-distributionDetection withTypicalFeatures
The mistaken rejection of an actually true null hypothesisH0 is the type I error.
- Information Technology > Artificial Intelligence > Representation & Reasoning (1.00)
- Information Technology > Artificial Intelligence > Machine Learning > Performance Analysis > Accuracy (0.95)
- Information Technology > Artificial Intelligence > Machine Learning > Neural Networks > Deep Learning > Generative AI (0.63)
YourOut-of-DistributionDetection MethodisNotRobust!
Although OOD detection methods have advanced by a great deal, they are still susceptible to adversarial examples, which is a violation of their purpose. To mitigate this issue, several defenses have recently been proposed. Nevertheless, these efforts remained ineffective, as their evaluations are based on either small perturbation sizes, or weak attacks.
