We introduce Joint Probability Trees (JPT), a novel approach that makes learning of and reasoning about joint probability distributions tractable for practical applications. JPTs support both symbolic and subsymbolic variables in a single hybrid model, and they do not rely on prior knowledge about variable dependencies or families of distributions. JPT representations build on tree structures that partition the problem space into relevant subregions that are elicited from the training data instead of postulating a rigid dependency model prior to learning. Learning and reasoning scale linearly in JPTs, and the tree structure allows white-box reasoning about any posterior probability $P(Q|E)$, such that interpretable explanations can be provided for any inference result. Our experiments showcase the practical applicability of JPTs in high-dimensional heterogeneous probability spaces with millions of training samples, making it a promising alternative to classic probabilistic graphical models.

翻译：我们提出联合概率树（JPT），这是一种新颖的方法，使得在实际应用中对联合概率分布的学习和推理变得可行。JPT在单个混合模型中同时支持符号变量和子符号变量，且不依赖关于变量依赖关系或分布族的先验知识。JPT表示基于树结构构建，这些树结构将问题空间划分为从训练数据中提取的相关子区域，而非在学习前预设刚性依赖模型。JPT中的学习和推理呈线性扩展，其树结构支持对任意后验概率$P(Q|E)$进行白箱推理，从而为任何推断结果提供可解释的阐述。我们的实验展示了JPT在高维异构概率空间（包含数百万训练样本）中的实际应用能力，使其成为经典概率图模型的有前景替代方案。