The familiarity hypothesis: Explaining the behavior of deep open set methods
作者：
Highlights：
• Open set detection using the maximum logit score of a softmax classifier matches the current state of the art.
• Familiarity Hypothesis: The Max Logit method detects the absence of familiarity rather than the presence of novelty.
• The reduced activity of positively-weighted object-relevant features accounts for most of the Max Logit score.
摘要
•Open set detection using the maximum logit score of a softmax classifier matches the current state of the art.•Familiarity Hypothesis: The Max Logit method detects the absence of familiarity rather than the presence of novelty.•The reduced activity of positively-weighted object-relevant features accounts for most of the Max Logit score.
论文关键词：Anomaly detection,Open set learning,Computer vision,Object recognition,Novel category detection,Representation learning,Deep learning
论文评审过程：Received 28 February 2022, Revised 24 June 2022, Accepted 21 July 2022, Available online 23 July 2022, Version of Record 31 July 2022.
论文官网地址：https://doi.org/10.1016/j.patcog.2022.108931
