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 2102.03509cs.LGstat.ML
收藏

使用Bernstein型多项式的鲁棒归一化流

Robust normalizing flows using Bernstein-type polynomials

Australian National University(澳大利亚国立大学);University of Toronto(多伦多大学);Mohamed bin Zayed University of AI(穆罕默德·本·扎耶德人工智能大学)

Sameera Ramasinghe, Kasun Fernando, Salman Khan, Nick Barnes

摘要:

由于样本数据会受到扰动,例如,仪器误差或添加的随机噪声。由于流模型是典型的非线性算法,它们放大了这些初始误差,导致泛化能力差。本文提出了一个框架来构建规范化流(NF),它表现出更高的鲁棒性,对这种初始错误。为此,我们利用伯恩斯坦型多项式的灵感来自伯恩斯坦基础的最佳稳定性。此外,与现有的NF框架相比,我们的方法提供了令人信服的优势,如近似误差的理论上限,更高的可解释性,适用于压缩支持的密度,以及在没有训练不稳定性的情况下采用更高次多项式的能力。我们进行了全面的理论分析,并通过在真实世界和合成数据集上的实验,实证证明了所提出的技术的有效性。

英文摘要:

Modeling real-world distributions can often be challenging due to sample data that are subjected to perturbations, e.g., instrumentation errors, or added random noise. Since flow models are typically nonlinear algorithms, they amplify these initial errors, leading to poor generalizations. This paper proposes a framework to construct Normalizing Flows (NF), which demonstrates higher robustness against such initial errors. To this end, we utilize Bernstein-type polynomials inspired by the optimal stability of the Bernstein basis. Further, compared to the existing NF frameworks, our method provides compelling advantages like theoretical upper bounds for the approximation error, higher interpretability, suitability for compactly supported densities, and the ability to employ higher degree polynomials without training instability. We conduct a thorough theoretical analysis and empirically demonstrate the efficacy of the proposed technique using experiments on both real-world and synthetic 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。

↑