arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2209.08183cs.LG
收藏

离散空间中局部平衡方案的最优尺度

Optimal Scaling for Locally Balanced Proposals in Discrete Spaces

Georgia Tech(佐治亚理工学院);Google Brain(谷歌大脑);U of Alberta(阿尔伯塔大学)

Haoran Sun, Hanjun Dai, Dale Schuurmans

摘要:

连续空间中的Metropolis-Hastings(M-H)算法的最优尺度已经得到了很好的研究,但在离散空间中缺乏类似的理解。最近,离散空间的一族局部平衡建议(LBP)已被证明是渐近最优的,但最优缩放的问题仍然是开放的。在本文中,我们建立,第一次,在离散空间中的M-H的效率也可以由一个渐近接受率,是独立的目标分布。此外,我们从理论和经验两方面验证了LBP和随机游走Metropolis(RWM)的最佳接受率分别为$0.574$和$0.234$。这些结果也有助于建立LBP是渐近$O(N^\frac{2}{3})$更有效的RWM方面的模型尺寸$N$。最优接受率的知识允许人们自动调整离散空间中建议分布的邻域大小,直接类似于连续空间中的步长控制。我们凭经验证明,这种自适应M-H采样可以鲁棒地改善离散空间中各种目标分布的采样,包括训练基于深度能量的模型。

英文摘要:

Optimal scaling has been well studied for Metropolis-Hastings (M-H) algorithms in continuous spaces, but a similar understanding has been lacking in discrete spaces. Recently, a family of locally balanced proposals (LBP) for discrete spaces has been proved to be asymptotically optimal, but the question of optimal scaling has remained open. In this paper, we establish, for the first time, that the efficiency of M-H in discrete spaces can also be characterized by an asymptotic acceptance rate that is independent of the target distribution. Moreover, we verify, both theoretically and empirically, that the optimal acceptance rates for LBP and random walk Metropolis (RWM) are $0.574$ and $0.234$ respectively. These results also help establish that LBP is asymptotically $O(N^\frac{2}{3})$ more efficient than RWM with respect to model dimension $N$. Knowledge of the optimal acceptance rate allows one to automatically tune the neighborhood size of a proposal distribution in a discrete space, directly analogous to step-size control in continuous spaces. We demonstrate empirically that such adaptive M-H sampling can robustly improve sampling in a variety of target distributions in discrete spaces, including training deep energy based models.

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。

↑