Belief Caching in 2APL

The BDI-oriented multi-agent programming language 2APL allows the implementation of an agent’s beliefs in terms of logical facts and rules. An agent’s beliefs represent information about its surrounding environment including other agents. Repeated querying of the beliefs by the 2APL interpreter causes unnecessary overhead resulting in poor runtime performance of the interpreter. We propose an extension to 2APL to reduce the number of such queries by using belief caching. We show that our proposal implements belief caching and extends an existing caching proposal. Moreover, we provide formal proofs establishing that our extension does not affect the execution behavior of 2APL. Benchmarking results indicate that belief caching leads to significant improvements.