Principled Detection of Out-of-Distribution Examples in Neural Networks

By Shiyu Liang, Yixuan Li, Rayadurgam Srikant. 2017

128 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Reliable Uncertainty Estimates in Deep Neural Networks using Noise Contrastive Priors (2018, methodology): “More recently, Lee et al. (2017) introduce a GAN to generate OOD samples, and Liang et al. (2018) add perturbations to the input, applying an “OOD detector” to improve softmax scores on OOD samples by scaling the temperature.”
Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samples (2017, methodology): “Specifically, (Hendrycks & Gimpel, 2016) defined the confidence score as a maximum value of the predictive distribution, and (Liang et al., 2017) further improved the performance by using temperature scaling (Guo et al., 2017) and adding small controlled perturbations to the input data.”
A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks (2018, methodology): “Here, we remark that abnormal samples can be characterized better in the representation space of DNNs, rather than the “label-overfitted” output space of softmax-based posterior distribution used in the prior works [13, 21] for detecting them.”
Dataset Inference: Ownership Resolution in Machine Learning (2021, background): “Liang et al. (2017) and Lee et al. (2018) measure model performance on modifying an input to ﬁnd if a sample is in or out-of-distribution.”
Noise Contrastive Priors for Functional Uncertainty (2018, methodology): “Recently, Lee et al. (2017) introduce a GAN to generate OOD samples, and Liang et al. (2018) add perturbations to the input, applying an “OOD detector” to improve softmax scores on OOD samples by scaling the temperature.”
Bayes-Probe: Distribution-Guided Sampling for Prediction Level Sets (2020, methodology): “While many methods have been proposed to detect out-of-distribution data (Vernekar et al., 2019; DeVries & Taylor, 2018; Hendrycks & Gimpel, 2016; Liang et al., 2017), BAYES-PROBE facilitates model insight by revealing examples from novel classes that are likely to cause a false-positive detection, ”
Detection and Retrieval of Out-of-Distribution Objects in Semantic Segmentation (2020, methodology): “The task of OOD detection has been broadly studied for image recognition [6,11,18,23] and most of these methods are applicable to the problem of semantic image segmentation.”