A0851
Title: Enhancing the power of OOD detection via sample-aware model selection
Authors: Chuanlong Xie - Beijing Normal University (China) [presenting]
Abstract: A novel perspective is presented on detecting out-of-distribution (OOD) samples and an algorithm is proposed for sample-aware model selection to enhance the effectiveness of OOD detection. The algorithm determines, for each test input, which pre-trained models in the model zoo are capable of identifying the test input as an OOD sample. If no such models exist in the model zoo, the test input is classified as an in-distribution (ID) sample. It is theoretically demonstrated that the method maintains the true positive rate of ID samples and accurately identifies OOD samples with high probability when there are a sufficient number of diverse pre-trained models in the model zoo. Extensive experiments were conducted to validate the method, demonstrating that it leverages the complementarity among single-model detectors to consistently improve the effectiveness of OOD sample identification. Compared to baseline methods, the approach improved the relative performance by 65.40\% and 37.25\% on the CIFAR10 and ImageNet benchmarks, respectively.
