A Unified Framework for Human-Allied Learning of Probabilistic Circuits
This work addresses the issue of knowledge-intensive learning in probabilistic circuits for domains such as healthcare, representing an incremental improvement by combining existing methods with domain knowledge.
The paper tackles the problem of learning probabilistic circuits by integrating domain knowledge into parameter learning, particularly in data-scarce domains like healthcare, and shows that the proposed framework achieves superior performance compared to purely data-driven methods.
Probabilistic Circuits (PCs) have emerged as an efficient framework for representing and learning complex probability distributions. Nevertheless, the existing body of research on PCs predominantly concentrates on data-driven parameter learning, often neglecting the potential of knowledge-intensive learning, a particular issue in data-scarce/knowledge-rich domains such as healthcare. To bridge this gap, we propose a novel unified framework that can systematically integrate diverse domain knowledge into the parameter learning process of PCs. Experiments on several benchmarks as well as real world datasets show that our proposed framework can both effectively and efficiently leverage domain knowledge to achieve superior performance compared to purely data-driven learning approaches.