arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.07564cs.LG
收藏

正则化和辅助信息在OOD检测中的重要性

On the Importance of Regularisation & Auxiliary Information in OOD Detection

University College Dublin(都柏林大学学院)

John Mitros, Brian Mac Namee

摘要:

神经网络通常用于关键领域的应用(例如自动驾驶汽车,金融市场和航空航天工程),即使它们对模糊的输入表现出过度自信的预测。这种缺陷表明了一个根本性的缺陷,即神经网络经常在虚假的相关性上过拟合。为了解决这个问题,在这项工作中,我们提出了两个新的目标,提高网络的能力,以检测出的分布样本,从而避免过度自信的预测模糊的输入。我们的经验表明,我们的方法优于基线和性能优于大多数现有的方法,同时仍然保持对其余的竞争力的表现。此外,我们经验证明了我们的方法对常见的腐败的鲁棒性,并证明了正则化和辅助信息在分布外检测的重要性。

英文摘要:

Neural networks are often utilised in critical domain applications (e.g. self-driving cars, financial markets, and aerospace engineering), even though they exhibit overconfident predictions for ambiguous inputs. This deficiency demonstrates a fundamental flaw indicating that neural networks often overfit on spurious correlations. To address this problem in this work we present two novel objectives that improve the ability of a network to detect out-of-distribution samples and therefore avoid overconfident predictions for ambiguous inputs. We empirically demonstrate that our methods outperform the baseline and perform better than the majority of existing approaches while still maintaining a competitive performance against the rest. Additionally, we empirically demonstrate the robustness of our approach against common corruptions and demonstrate the importance of regularisation and auxiliary information in out-of-distribution detection.

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。

↑