arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1902.01893cs.LGstat.ML
收藏

TzK:基于流的条件生成模型

TzK: Flow-Based Conditional Generative Model

University of Toronto(多伦多大学);Vector Institute(向量研究所)

Micha Livne, David Fleet

摘要:

我们制定了一类新的基于概率流的条件生成模型。通过最大似然训练,它可以从类条件或联合分布中提供有效的推理和采样,并且不需要类的数量或类之间的关系的先验知识。这允许人们从多个异构数据集训练生成模型,同时在数据子集上保留强先验模型(例如,从单个数据集、类标签或属性)。在本文中,除了端到端学习之外,我们还展示了如何从具有相对较弱的Glow架构的多个数据集中学习单个模型,然后通过调节不同的知识类型(例如,单个数据集)。这产生了与最先进的、令人信服的条件先验样本相当的对数似然。

英文摘要:

We formulate a new class of conditional generative models based on probability flows. Trained with maximum likelihood, it provides efficient inference and sampling from class-conditionals or the joint distribution, and does not require a priori knowledge of the number of classes or the relationships between classes. This allows one to train generative models from multiple, heterogeneous datasets, while retaining strong prior models over subsets of the data (e.g., from a single dataset, class label, or attribute). In this paper, in addition to end-to-end learning, we show how one can learn a single model from multiple datasets with a relatively weak Glow architecture, and then extend it by conditioning on different knowledge types (e.g., a single dataset). This yields log likelihood comparable to state-of-the-art, compelling samples from conditional priors.

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。

↑