Finding Structure and Causality in Linear Programs
This work addresses the need for deeper understanding of LPs in machine learning, potentially enhancing inference and structured learning systems, but appears incremental as it builds on existing LP frameworks.
The authors tackled the problem of uncovering structure and causality in Linear Programs (LPs), revealing intra- and inter-structure relations through a foundational, causal perspective, with empirical investigation on general, shortest path, and energy system LPs.
Linear Programs (LP) are celebrated widely, particularly so in machine learning where they have allowed for effectively solving probabilistic inference tasks or imposing structure on end-to-end learning systems. Their potential might seem depleted but we propose a foundational, causal perspective that reveals intriguing intra- and inter-structure relations for LP components. We conduct a systematic, empirical investigation on general-, shortest path- and energy system LPs.