2013/02/20 by Sampath Srinivas, Eric Horvitz, Srinivas, Sampath +1 · 1 citation
Computer Science · #AI-based Problem Solving and Planning #Artificial Intelligence (cs.AI) #Bayesian Modeling and Causal Inference #FOS: Computer and information sciences #Machine Learning and Algorithms
paper · pdf · doi:10.48550/arxiv.1302.4986
openalex publication_date 2013/02/20 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
The goal of model-based diagnosis is to isolate causes of anomalous system\nbehavior and recommend inexpensive repair actions in response. In general,\nprecomputing optimal repair policies is intractable. To date, investigators\naddressing this problem have explored approximations that either impose\nrestrictions on the system model (such as a single fault assumption) or compute\nan immediate best action with limited lookahead. In this paper, we develop a\nformulation of repair in model-based diagnosis and a repair algorithm that\ncomputes optimal sequences of actions. This optimal approach is costly but can\nbe applied to precompute an optimal repair strategy for compact systems. We\nshow how we can exploit a hierarchical system specification to make this\napproach tractable for large systems. When introducing hierarchy, we also\nconsider the tradeoff between simply replacing a component and decomposing it\nto repair its subcomponents. The hierarchical repair algorithm is suitable for\noff-line precomputation of an optimal repair strategy. A modification of the\nalgorithm takes advantage of an iterative deepening scheme to trade off\ninference time and the quality of the computed strategy.\n