arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2012.12687cs.LGstat.ML
收藏

瓦瑟斯坦辍学

Wasserstein Dropout

Fraunhofer IAIS(弗劳恩霍夫智能分析与信息系统研究所);Fraunhofer Center for Machine Learning(弗劳恩霍夫机器学习中心);University of Bonn(波恩大学);University of Bochum(波鸿大学)

Joachim Sicking, Maram Akila, Maximilian Pintz, Tim Wirtz, Asja Fischer, Stefan Wrobel

摘要:

尽管对安全的机器学习具有重要意义,但神经网络的不确定性量化问题还远未解决。估计神经不确定性的最新方法通常是混合的,将参数模型与显式或隐式(基于辍学)集成相结合。我们采取另一种途径,并提出了一种新的方法,不确定性量化回归任务,Wasserstein辍学,这是纯粹的非参数。从技术上讲,它通过基于丢弃的子网络分布来捕获任意的不确定性。这是通过一个新的目标,最大限度地减少标签分布和模型分布之间的Wasserstein距离。一个广泛的实证分析表明,Wasserstein dropout优于最先进的方法,香草测试数据以及下的分布转移,在产生更准确和稳定的不确定性估计。

英文摘要:

Despite of its importance for safe machine learning, uncertainty quantification for neural networks is far from being solved. State-of-the-art approaches to estimate neural uncertainties are often hybrid, combining parametric models with explicit or implicit (dropout-based) ensembling. We take another pathway and propose a novel approach to uncertainty quantification for regression tasks, Wasserstein dropout, that is purely non-parametric. Technically, it captures aleatoric uncertainty by means of dropout-based sub-network distributions. This is accomplished by a new objective which minimizes the Wasserstein distance between the label distribution and the model distribution. An extensive empirical analysis shows that Wasserstein dropout outperforms state-of-the-art methods, on vanilla test data as well as under distributional shift, in terms of producing more accurate and stable uncertainty estimates.

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。

↑