arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1906.03260stat.MLcs.LG
收藏

方差网络的可靠训练和估计

Reliable training and estimation of variance networks

Technical University of Denmark(丹麦技术大学)

Nicki S. Detlefsen, Martin Jørgensen, Søren Hauberg

摘要:

我们提出并研究了新的补充方法,估计预测方差网络回归神经网络。我们推导出一个局部感知的mini-slave方案,该方案产生稀疏的鲁棒梯度,并展示了如何对方差网络进行无偏权重更新。此外,我们制定了一个启发式的鲁棒拟合的均值和方差网络事后。最后,我们从后验高斯过程中得到启发,提出了一种具有类似高斯过程外推特性的网络结构。所提议的方法是互补的,并分别改进了基线方法。在实验中,我们研究了预测不确定性对多个数据集和任务的影响,包括回归,主动学习和生成建模。实验一致表明,与跨任务和数据集的最先进方法相比,预测不确定性估计有显着改进。

英文摘要:

We propose and investigate new complementary methodologies for estimating predictive variance networks in regression neural networks. We derive a locally aware mini-batching scheme that result in sparse robust gradients, and show how to make unbiased weight updates to a variance network. Further, we formulate a heuristic for robustly fitting both the mean and variance networks post hoc. Finally, we take inspiration from posterior Gaussian processes and propose a network architecture with similar extrapolation properties to Gaussian processes. The proposed methodologies are complementary, and improve upon baseline methods individually. Experimentally, we investigate the impact on predictive uncertainty on multiple datasets and tasks ranging from regression, active learning and generative modeling. Experiments consistently show significant improvements in predictive uncertainty estimation over state-of-the-art methods across tasks and datasets.

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。

↑