Concetti Chiave
This paper proposes four forms of agentive permissions in multiagent settings: strong permission to admit, strong permission to ensure, weak permission to admit, and weak permission to ensure. It provides a formal semantics, analyzes the complexity of model checking, and presents a sound and complete logical system capturing the interplay between these modalities.
Sintesi
The paper introduces four forms of agentive permissions in multiagent settings:
Strong permission to admit (SA): Any action of an agent that admits of a certain outcome is permitted.
Strong permission to ensure (SE): Any action of an agent that ensures a certain outcome is permitted.
Weak permission to admit (WA): There exists a permitted action of an agent that admits of a certain outcome.
Weak permission to ensure (WE): There exists a permitted action of an agent that ensures a certain outcome.
The paper provides a formal semantics for these modalities in multiagent transition systems. It analyzes the time complexity of the model checking problem, showing it is polynomial in the size of the input. The paper also proves that these four modalities are semantically undefinable through each other, in contrast to the fact that "ensuring" and "admitting" are dual notions when separated from permissions.
Finally, the paper presents a sound and complete logical system for reasoning about these four forms of agentive permissions. The axioms and inference rules capture the interplay between the modalities and offer an efficient way for permission reasoning in multiagent settings.