A Learning-Based Algorithm Selection Meta-reasoner for the Real-Time MPE Problem

Abstract. The algorithm selection problem aims to select the best algorithm for an input problem instance according to some characteristics of the instance. This paper presents a learning-based inductive approach to build a predictive algorithm selection system from empirical algorithm performance data of the Most Probable Explanation(MPE) problem. The learned model can serve as an algorithm selection meta-reasoner for the real-time MPE problem. Experimental results show that the learned algorithm selection models can help integrate multiple MPE algorithms to gain a better overall performance of reasoning.

• Type: Conference
• Year: 2004
• Where: AUSAI
• Authors: Haipeng Guo, William H. Hsu
• Keywords: Algorithm Selection, Algorithm Selection Problem, Artificial Intelligence, AUSAI 2004, Predictive Algorithm Selection