A Pragmatic and Semantic Unified Framework for Agent Communication (original) (raw)

A modal semantics for an argumentation-based pragmatics for agent communication

2005

Abstract In this paper we present a modal semantics for our approach based on social commitments and arguments for conversational agents. Our formal framework based on this approach uses three basic elements: social commitments, actions that agents apply to these social commitments and arguments that agents use to support their actions. This framework, called Commitment and Argument Network (CAN), formalizes the agents' interactions as a network in which agents manipulate commitments and arguments.

Towards a formal framework for conversational agents

2003

ABSTRACT This paper proposes a formal framework which offers an external representation of conversations between conversational agents. Using this formalism allows us:(1) to represent the dynamics of conversations between agents;(2) to analyze conversations;(3) to help autonomous agents to take part in consistent conversations. The proposed formalism, called “commitment and argument network”, uses a combined approach based on commitments and arguments.

Rules for Formal and Natural Dialogues in Agent Communication

2012

The paper aims to bring together and unify two traditions in studying dialogue as a game: dialogical logic introduced by Lorenzen; and persuasion dialogue games as specified by Prakken. The first approach allows the representation of formal dialogues in which the validity of argument is the topic discussed. The second tradition has focused on natural dialogues examining, e.g., informal fallacies typical in real-life communication. Our goal is to unite these two approaches in order to allow communicating agents to benefit from the advantages of both, i.e. to equip them with the ability to persuade each other not only about facts, but also about the classical propositional validity of argument used in a dialogue. To this end, Lorenzen’s system needs to be expressed according to the generic specification for natural dialogues proposed by Prakken. As a result, the system proposed in the paper allows the representation and elimination of formal fallacies committed during a dialogue.

A Framework for Agent Communication based on Goals and Argumentation

2008

This paper proposes a formal framework for agent communication where agents can reason about their goals using strategic reasoning. This reasoning is argumentation-based and enables agents to generate a set of strategic goals depending on a set of constraints. Sub-goals are generated using this reasoning and they can be cancelled or substituted for alternatives during the dialogue progress. An original characteristic of this framework is that agents can use this strategic reasoning together with a tactic reasoning to persist in the achievement of their goals by considering alternatives depending on a set of constraints. Tactic reasoning is responsible of selecting the communicative acts to perform in order to realize the strategic goals. Some constraints are fixed when the conversation starts and others during the dialogue progress. The paper also discusses the computational complexity of such a reasoning.

The mechanics of some formal inter-agent dialogues

2004

This paper studies argumentation-based dialogues between agents. It takes a previously defined system by which agents can trade arguments and examines in detail what locutions are passed between agents. This makes it possible to identify finer-grained protocols than has been previously possible, exposing the relationships between different kinds of dialogue, and giving a deeper understanding of how such dialogues could be automated.

On the relevance of utterances in formal inter-agent dialogues

2007

Abstract Work on argumentation-based dialogue has defined frameworks within which dialogues can be carried out, established protocols that govern dialogues, and studied different properties of dialogues. This work has established the space in which agents are permitted to interact through dialogues. Recently, there has been increasing interest in the mechanisms agents might use to choose how to act---the rhetorical manoeuvring that they use to navigate through the space defined by the rules of the dialogue.

An argumentation-based semantics for agent communication languages

2002

Abstract. In recent years, the importance of defining a standard framework for agent communication language (ACL) has been widely recognized. However, classical proposals (mentalistic semantics and social semantics) fail to meet the objectives of verifiability and flexibility required in complex interactions involving heterogenous agents possibly designed by different programmers.