arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1905.03319cs.LGstat.ML
收藏

数据有效的互信息神经网络估计

Data-Efficient Mutual Information Neural Estimator

SRI International(SRI国际);Princeton University(普林斯顿大学)

Xiao Lin, Indranil Sur, Samuel A. Nastase, Ajay Divakaran, Uri Hasson, Mohamed R. Amer

摘要:

从观测样本中测量高维连续随机变量之间的互信息具有广泛的理论和实际应用。最近的工作MINE(Belghazi et al. 2018)专注于使用神经网络估计MI的紧变分下界,但假设样本供应无限以防止过拟合。在现实世界的应用程序中,数据并不总是过剩的。在这项工作中,我们专注于提高数据效率,并提出了一个数据有效的矿估计(DEMINE),通过开发一个宽松的预测MI下限,可以估计在更高的数据效率的数量级。预测MI下限还使我们能够开发一种新的元学习方法,使用任务增强Meta-DEMINE,以提高网络的泛化能力,并进一步提高经验估计精度。随着数据效率的提高,我们的估计器可以在实际的数据集大小上进行依赖性的统计测试。我们证明了我们的估计的有效性合成基准和现实世界的功能磁共振成像数据,与学科间的相关性分析的应用。

英文摘要:

Measuring Mutual Information (MI) between high-dimensional, continuous, random variables from observed samples has wide theoretical and practical applications. Recent work, MINE (Belghazi et al. 2018), focused on estimating tight variational lower bounds of MI using neural networks, but assumed unlimited supply of samples to prevent overfitting. In real world applications, data is not always available at a surplus. In this work, we focus on improving data efficiency and propose a Data-Efficient MINE Estimator (DEMINE), by developing a relaxed predictive MI lower bound that can be estimated at higher data efficiency by orders of magnitudes. The predictive MI lower bound also enables us to develop a new meta-learning approach using task augmentation, Meta-DEMINE, to improve generalization of the network and further boost estimation accuracy empirically. With improved data-efficiency, our estimators enables statistical testing of dependency at practical dataset sizes. We demonstrate the effectiveness of our estimators on synthetic benchmarks and a real world fMRI data, with application of inter-subject correlation analysis.

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。

↑