{"title":"面部动作单元强度回归中不确定性预测的简单有效方法。","authors":"Torsten Wörtwein, Louis-Philippe Morency","doi":"10.1109/fg47880.2020.00045","DOIUrl":null,"url":null,"abstract":"<p><p>Knowing how much to trust a prediction is important for many critical applications. We describe two simple approaches to estimate uncertainty in regression prediction tasks and compare their performance and complexity against popular approaches. We operationalize uncertainty in regression as the absolute error between a model's prediction and the ground truth. Our two proposed approaches use a secondary model to predict the uncertainty of a primary predictive model. Our first approach leverages the assumption that similar observations are likely to have similar uncertainty and predicts uncertainty with a non-parametric method. Our second approach trains a secondary model to directly predict the uncertainty of the primary predictive model. Both approaches outperform other established uncertainty estimation approaches on the MNIST, DISFA, and BP4D+ datasets. Furthermore, we observe that approaches that directly predict the uncertainty generally perform better than approaches that indirectly estimate uncertainty.</p>","PeriodicalId":87341,"journal":{"name":"Proceedings of the ... International Conference on Automatic Face and Gesture Recognition. IEEE International Conference on Automatic Face & Gesture Recognition","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1109/fg47880.2020.00045","citationCount":"3","resultStr":"{\"title\":\"Simple and Effective Approaches for Uncertainty Prediction in Facial Action Unit Intensity Regression.\",\"authors\":\"Torsten Wörtwein, Louis-Philippe Morency\",\"doi\":\"10.1109/fg47880.2020.00045\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Knowing how much to trust a prediction is important for many critical applications. We describe two simple approaches to estimate uncertainty in regression prediction tasks and compare their performance and complexity against popular approaches. We operationalize uncertainty in regression as the absolute error between a model's prediction and the ground truth. Our two proposed approaches use a secondary model to predict the uncertainty of a primary predictive model. Our first approach leverages the assumption that similar observations are likely to have similar uncertainty and predicts uncertainty with a non-parametric method. Our second approach trains a secondary model to directly predict the uncertainty of the primary predictive model. Both approaches outperform other established uncertainty estimation approaches on the MNIST, DISFA, and BP4D+ datasets. Furthermore, we observe that approaches that directly predict the uncertainty generally perform better than approaches that indirectly estimate uncertainty.</p>\",\"PeriodicalId\":87341,\"journal\":{\"name\":\"Proceedings of the ... International Conference on Automatic Face and Gesture Recognition. IEEE International Conference on Automatic Face & Gesture Recognition\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://sci-hub-pdf.com/10.1109/fg47880.2020.00045\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the ... International Conference on Automatic Face and Gesture Recognition. IEEE International Conference on Automatic Face & Gesture Recognition\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/fg47880.2020.00045\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2021/1/18 0:00:00\",\"PubModel\":\"Epub\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the ... International Conference on Automatic Face and Gesture Recognition. IEEE International Conference on Automatic Face & Gesture Recognition","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/fg47880.2020.00045","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2021/1/18 0:00:00","PubModel":"Epub","JCR":"","JCRName":"","Score":null,"Total":0}
Simple and Effective Approaches for Uncertainty Prediction in Facial Action Unit Intensity Regression.
Knowing how much to trust a prediction is important for many critical applications. We describe two simple approaches to estimate uncertainty in regression prediction tasks and compare their performance and complexity against popular approaches. We operationalize uncertainty in regression as the absolute error between a model's prediction and the ground truth. Our two proposed approaches use a secondary model to predict the uncertainty of a primary predictive model. Our first approach leverages the assumption that similar observations are likely to have similar uncertainty and predicts uncertainty with a non-parametric method. Our second approach trains a secondary model to directly predict the uncertainty of the primary predictive model. Both approaches outperform other established uncertainty estimation approaches on the MNIST, DISFA, and BP4D+ datasets. Furthermore, we observe that approaches that directly predict the uncertainty generally perform better than approaches that indirectly estimate uncertainty.