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 2210.00993cs.LGcs.AIstat.ML
收藏

通过自适应重要性采样实现神经网络中的高效贝叶斯推断

Efficient Bayes Inference in Neural Networks through Adaptive Importance Sampling

ETS Montréal(蒙特利尔高等技术学院);Inria Saclay(法国国家信息与自动化研究所萨克莱分所);CentraleSupélec(巴黎中央理工-高等电力学院);Université Paris-Saclay(巴黎萨克雷大学);University of Edinburgh(爱丁堡大学)

Yunshi Huang, Emilie Chouzenoux, Victor Elvira, Jean-Christophe Pesquet

AI总结:

本文提出PMCnet算法,将自适应重要性采样与后验分布几何信息相结合,用于高效训练贝叶斯神经网络,在浅层和深层网络上均表现出优异的性能与探索能力。

摘要:

贝叶斯神经网络(BNNs)近年来受到越来越多的关注。在BNNs中,网络未知权重和偏置参数的完整后验分布在训练阶段被生成。这种概率估计相较于逐点估计具有若干优势,特别是能够在预测新数据时提供不确定性量化。这一贝叶斯范式固有的特性在无数机器学习应用中十分有用,在决策具有关键影响的领域尤其具有吸引力,例如医疗健康或自动驾驶。BNNs的主要挑战在于训练过程的计算成本,因为贝叶斯技术常常面临严重的维度灾难。自适应重要性采样(AIS)是最重要的蒙特卡洛方法之一,具有可靠的收敛保证和易于自适应的优点。本工作旨在表明AIS构成了设计BNNs的一种成功方法。更准确地说,我们提出了一种新算法PMCnet,该算法包含一种高效的自适应机制,利用复杂(通常是多模态)后验分布的几何信息。数值结果展示了所提方法在浅层和深层神经网络上均具有出色的性能和更强的探索能力。

英文摘要:

Bayesian neural networks (BNNs) have received an increased interest in the last years. In BNNs, a complete posterior distribution of the unknown weight and bias parameters of the network is produced during the training stage. This probabilistic estimation offers several advantages with respect to point-wise estimates, in particular, the ability to provide uncertainty quantification when predicting new data. This feature inherent to the Bayesian paradigm, is useful in countless machine learning applications. It is particularly appealing in areas where decision-making has a crucial impact, such as medical healthcare or autonomous driving. The main challenge of BNNs is the computational cost of the training procedure since Bayesian techniques often face a severe curse of dimensionality. Adaptive importance sampling (AIS) is one of the most prominent Monte Carlo methodologies benefiting from sounded convergence guarantees and ease for adaptation. This work aims to show that AIS constitutes a successful approach for designing BNNs. More precisely, we propose a novel algorithm PMCnet that includes an efficient adaptation mechanism, exploiting geometric information on the complex (often multimodal) posterior distribution. Numerical results illustrate the excellent performance and the improved exploration capabilities of the proposed method for both shallow and deep neural networks.

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。

↑