Explaining Deep Tractable Probabilistic Models: The sum-product network case
arXiv:2110.09778
Abstract
We consider the problem of explaining a class of tractable deep probabilistic models, the Sum-Product Networks (SPNs) and present an algorithm ExSPN to generate explanations. To this effect, we define the notion of a context-specific independence tree(CSI-tree) and present an iterative algorithm that converts an SPN to a CSI-tree. The resulting CSI-tree is both interpretable and explainable to the domain expert. We achieve this by extracting the conditional independencies encoded by the SPN and approximating the local context specified by the structure of the SPN. Our extensive empirical evaluations on synthetic, standard, and real-world clinical data sets demonstrate that the CSI-tree exhibits superior explainability.
Main paper: 8 pages, references: 1 page. Main paper: 4 figures
References in corpus (8)
- Context-Specific Independence in Bayesian Networks
- Sum-Product Networks: A New Deep Architecture
- Deep Probabilistic Programming
- Random Feature Expansions for Deep Gaussian Processes
- SPFlow: An Easy and Extensible Library for Deep Probabilistic Learning using Sum-Product Networks
- On Relaxing Determinism in Arithmetic Circuits
- Identifying Causal Effects via Context-specific Independence Relations
- A New Perspective on Learning Context-Specific Independence