arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2303.04115cs.CVcs.LG
收藏

大规模分布外检测的预测嵌入功效回归

Predicted Embedding Power Regression for Large-Scale Out-of-Distribution Detection

Rochester Institute of Technology(罗切斯特理工学院)

Hong Yang, William Gebhardt, Alexander G. Ororbia, Travis Desell

摘要:

分发外(OOD)输入可能会损害现实世界机器学习系统的性能和安全性。虽然存在许多用于OOD检测的方法,并且在具有较低分辨率和较少类别的小规模数据集上工作良好,但是很少有方法被开发用于大规模OOD检测。现有的大规模方法通常依赖于最大分类概率,例如最先进的分组softmax方法。在这项工作中,我们开发了一种新的方法,该方法基于在训练过程中学习到的标签分布来计算预测类标签的概率。我们的方法比当前最先进的方法性能更好,计算成本只有微不足道的增加。我们在$14$数据集上对我们的方法进行了评估,并在AUROC(84.2 vs 82.4)和AUPR(96.2 vs 93.7)方面取得了统计学显著的改善。

英文摘要:

Out-of-distribution (OOD) inputs can compromise the performance and safety of real world machine learning systems. While many methods exist for OOD detection and work well on small scale datasets with lower resolution and few classes, few methods have been developed for large-scale OOD detection. Existing large-scale methods generally depend on maximum classification probability, such as the state-of-the-art grouped softmax method. In this work, we develop a novel approach that calculates the probability of the predicted class label based on label distributions learned during the training process. Our method performs better than current state-of-the-art methods with only a negligible increase in compute cost. We evaluate our method against contemporary methods across $14$ datasets and achieve a statistically significant improvement with respect to AUROC (84.2 vs 82.4) and AUPR (96.2 vs 93.7).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑