arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1504.02526cs.LGcs.DS
收藏

学习离散分布的任意统计混合

Learning Arbitrary Statistical Mixtures of Discrete Distributions

Tsinghua University(清华大学);The Hebrew University of Jerusalem(耶路撒冷希伯来大学);California Institute of Technology(加州理工学院)

Jian Li, Yuval Rabani, Leonard J. Schulman, Chaitanya Swamy

摘要:

本文研究了大型有限集上的非标记样本的非常一般的统计混合模型的学习问题。具体地,要学习的模型$\vartheta$是概率分布$p$上的概率分布,其中每个这样的$p$是$[n] = \{1,2,\dots,n\}$上的概率分布。当我们从$\vartheta$采样时,我们不直接观察$p$,而只是通过从分布$p$重复地、独立地从$[n]$采样$K$次来间接地并且以非常嘈杂的方式观察$p$。问题是在运输(推土机)距离中高精度地推断$\vartheta$。我们给出了第一个有效的算法来学习这个混合模型,而不对分布$\vartheta$的结构作任何限制性假设。我们将解决方案的质量作为样本大小$K$和所用样本数量的函数。我们的模型和结果应用于各种无监督学习场景,包括学习主题模型和协同过滤。

英文摘要:

We study the problem of learning from unlabeled samples very general statistical mixture models on large finite sets. Specifically, the model to be learned, $\vartheta$, is a probability distribution over probability distributions $p$, where each such $p$ is a probability distribution over $[n] = \{1,2,\dots,n\}$. When we sample from $\vartheta$, we do not observe $p$ directly, but only indirectly and in very noisy fashion, by sampling from $[n]$ repeatedly, independently $K$ times from the distribution $p$. The problem is to infer $\vartheta$ to high accuracy in transportation (earthmover) distance. We give the first efficient algorithms for learning this mixture model without making any restricting assumptions on the structure of the distribution $\vartheta$. We bound the quality of the solution as a function of the size of the samples $K$ and the number of samples used. Our model and results have applications to a variety of unsupervised learning scenarios, including learning topic models and collaborative filtering.

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。

↑