OSLฮฑ: Online Structure Learning using Background Knowledge Axiomatization


We present ๐™พ๐š‚๐™ป๐›ผโ€”an online structure learner for Markov Logic Networks (MLNs) that exploits background knowledge axiomatization in order to constrain the space of possible structures. Many domains of interest are characterized by uncertainty and complex relational structure. MLNs is a state-of-the-art Statistical Relational Learning framework that can naturally be applied to domains governed by these characteristics. Learning MLNs from data is challenging, as their relational structure increases the complexity of the learning process. In addition, due to the dynamic nature of many real-world applications, it is desirable to incrementally learn or revise the modelโ€™s structure and parameters. Experimental results are presented in activity recognition using a probabilistic variant of the Event Calculus (๐™ผ๐™ป๐™ฝโˆ’๐™ด๐™ฒ) as background knowledge and a benchmark dataset for video surveillance.

In European Conference on Machine Learning
Alexander Artikis
Alexander Artikis
Associate Professor of Artificial Intelligence