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 2009.09535stat.MLcs.LG
收藏

具有自适应漂移的随机梯度Langevin动力学算法

Stochastic Gradient Langevin Dynamics Algorithms with Adaptive Drifts

Purdue University(普渡大学)

Sehwan Kim, Qifan Song, Faming Liang

摘要:

贝叶斯深度学习提供了一种解决人工智能(AI)安全性问题的原则性方法,例如模型不确定性,模型可解释性和预测偏差。然而,由于缺乏从深度神经网络(DNN)的后验中采样的有效蒙特卡罗算法,贝叶斯深度学习尚未为我们的AI系统提供动力。提出了一类自适应随机梯度马尔可夫链蒙特卡罗(SGMCMC)算法,该算法对漂移函数进行偏置以增强对鞍点的逃逸,并根据过去样本的梯度自适应地调整偏置.我们建立了所提出的算法在温和的条件下的收敛性,并通过数值例子表明,所提出的算法可以显着优于现有的SGMCMC算法,如随机梯度朗之万动力学(SGLD),随机梯度哈密顿蒙特卡罗(SGHMC)和预处理SGLD,在模拟和优化任务。

英文摘要:

Bayesian deep learning offers a principled way to address many issues concerning safety of artificial intelligence (AI), such as model uncertainty,model interpretability, and prediction bias. However, due to the lack of efficient Monte Carlo algorithms for sampling from the posterior of deep neural networks (DNNs), Bayesian deep learning has not yet powered our AI system. We propose a class of adaptive stochastic gradient Markov chain Monte Carlo (SGMCMC) algorithms, where the drift function is biased to enhance escape from saddle points and the bias is adaptively adjusted according to the gradient of past samples. We establish the convergence of the proposed algorithms under mild conditions, and demonstrate via numerical examples that the proposed algorithms can significantly outperform the existing SGMCMC algorithms, such as stochastic gradient Langevin dynamics (SGLD), stochastic gradient Hamiltonian Monte Carlo (SGHMC) and preconditioned SGLD, in both simulation and optimization tasks.

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。

↑