arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.08892cs.CV
收藏

基于不确定性动量模型的无监督嵌入学习

Unsupervised Embedding Learning from Uncertainty Momentum Modeling

Northwestern University(西北大学);University of Oxford(牛津大学);Renmin University of China(中国人民大学)

Jiahuan Zhou, Yansong Tang, Bing Su, Ying Wu

摘要:

现有流行的无监督嵌入学习方法主要通过探索各种负面数据来增强给定未标记图像的实例级局部区分度。然而,存在的样本离群值表现出大的类内分歧或小的类间变化严重限制了他们的学习性能。我们证明性能限制是由梯度消失在这些样本离群值。此外,缺乏积极的数据和忽视全局歧视的考虑也提出了关键问题的无监督学习,但总是被忽视的现有方法。为了解决这些问题,我们提出了一种新的解决方案来显式建模并直接探索给定未标记学习样本的不确定性。而不是学习一个确定性的特征点为每个样本在嵌入空间中,我们建议表示一个样本的随机高斯与平均向量描绘其空间定位和协方差向量表示样本的不确定性。我们利用这种不确定性建模作为学习的动力,这有助于解决离群值。此外,可以容易地从所学习的实例特定的分布中提取大量的正候选,其被进一步采用以减轻上述问题。深入的理论分析和广泛的实验,以验证我们的优越性。

英文摘要:

Existing popular unsupervised embedding learning methods focus on enhancing the instance-level local discrimination of the given unlabeled images by exploring various negative data. However, the existed sample outliers which exhibit large intra-class divergences or small inter-class variations severely limit their learning performance. We justify that the performance limitation is caused by the gradient vanishing on these sample outliers. Moreover, the shortage of positive data and disregard for global discrimination consideration also pose critical issues for unsupervised learning but are always ignored by existing methods. To handle these issues, we propose a novel solution to explicitly model and directly explore the uncertainty of the given unlabeled learning samples. Instead of learning a deterministic feature point for each sample in the embedding space, we propose to represent a sample by a stochastic Gaussian with the mean vector depicting its space localization and covariance vector representing the sample uncertainty. We leverage such uncertainty modeling as momentum to the learning which is helpful to tackle the outliers. Furthermore, abundant positive candidates can be readily drawn from the learned instance-specific distributions which are further adopted to mitigate the aforementioned issues. Thorough rationale analyses and extensive experiments are presented to verify our superiority.

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。

↑