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 1912.03488cs.LGstat.ML
收藏

标签噪声下的鲁棒深度有序回归

Robust Deep Ordinal Regression Under Label Noise

IIIT Hyderabad(印度理工学院海得拉巴分校)

Bhanu Garg, Naresh Manwani

摘要:

真实世界的数据往往容易受到标签噪声的影响,这可能会限制现有的有序回归算法的有效性。现有的有序回归工作不考虑标签噪声。我们提出了一个理论上接地的方法,类条件标签噪声在有序回归问题。我们提出了一个深度学习实现,用于顺序回归的两个常用损失函数,它既对标签噪声具有鲁棒性,又对良好的排名规则具有排名一致性。我们验证了这些属性的算法经验和鲁棒性标签噪声的真实数据和排名的一致性。据我们所知,这是稳健有序回归模型的第一种方法。

英文摘要:

The real-world data is often susceptible to label noise, which might constrict the effectiveness of the existing state of the art algorithms for ordinal regression. Existing works on ordinal regression do not take label noise into account. We propose a theoretically grounded approach for class conditional label noise in ordinal regression problems. We present a deep learning implementation of two commonly used loss functions for ordinal regression that is both - 1) robust to label noise, and 2) rank consistent for a good ranking rule. We verify these properties of the algorithm empirically and show robustness to label noise on real data and rank consistency. To the best of our knowledge, this is the first approach for robust ordinal regression models.

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。

↑