Out-of-distribution (OOD) detection is a crucial part of deploying machine learning models safely. It has been extensively studied with a plethora of methods developed in the literature. This problem is tackled with an OOD score computation, however, previous methods compute the OOD scores with limited usage of the in-distribution dataset. For instance, the OOD scores are computed with information from a small portion of the in-distribution data. Furthermore, these methods encode images with a neural image encoder. The robustness of these methods is rarely checked with respect to image encoders of different training methods and architectures. In this work, we introduce the diffusion process into the OOD task. The diffusion model integrates information on the whole training set into the predicted noise vectors. What's more, we deduce a closed-form solution for the noise vector (stable point). Then the noise vector is converted into our OOD score, we test both the deep model predicted noise vector and the closed-form noise vector on the OOD benchmarks \cite{openood}. Our method outperforms previous OOD methods across all types of image encoders (Table. \ref{main}). A $3.5\%$ performance gain is achieved with the MAE-based image encoder. Moreover, we studied the robustness of OOD methods by applying different types of image encoders. Some OOD methods failed to generalize well when switching image encoders from ResNet to Vision Transformers, our method performs exhibits good robustness with all the image encoders.
NoiseRater: Meta-Learned Noise Valuation for Diffusion Model Training
Stanford University(斯坦福大学);UNSW(新南威尔士大学);UCL(伦敦大学学院);The University of Chicago(芝加哥大学);CUHK(香港中文大学);Nanjing University(南京大学);Brown University(布朗大学);Yale University(耶鲁大学);University of Notre Dame(Notre Dame 大学);University of Waterloo(滑铁卢大学);UCB(加州大学伯克利分校);Georgia Technology(佐治亚理工学院);Amazon(亚马逊)
Diffusion^2: Dual Diffusion Model with Uncertainty-Aware Adaptive Noise for Momentary Trajectory Prediction
Civil and Environmental Engineering Department, University of Wisconsin-Madison(威斯康星大学麦迪逊分校土木与环境工程系);Civil and Environmental Engineering Department, University of Washington(华盛顿大学土木与环境工程系);College of Transportation, Tongji University(同济大学交通运输学院)
Sequence-Adaptive Video Prediction in Continuous Streams using Diffusion Noise Optimization
University of Bonn(波恩大学);Toyota Motor Europe(丰田欧洲公司);University of Wuerzburg(乌尔姆大学);Lamarr Institute for Machine Learning and Artificial Intelligence(拉马尔人工智能与机器学习研究所)
Energy-Conditioned Noise Schedule and Whitening for Spectral Diffusion
University of Nis, Faculty of Electronic Engineering(尼什大学电子工程学院);University MB, Information Technology department(MB大学信息技术学院);University of Arizona, Electrical & Computer Engineering (ECE) department(亚利桑那大学电气与计算机工程系)
NoiseDiffusion: Correcting Noise for Image Interpolation with Diffusion Models beyond Spherical Linear Interpolation
University of Science and Technology of China(中国科学技术大学);Hong Kong Baptist University(香港浸会大学);University of Technology Sydney(悉尼科技大学);The University of Sydney(悉尼大学);Sydney AI Centre(悉尼人工智能中心)