Communicating Inductive Inferences In Agent Societies
نویسندگان
چکیده
Our work was motivated by a lack of support for machine learning capabilities in agent communication languages such as KQML, Agent-0 and FIPA. The requirements were that such any extension to these communication languages should allow heterogeneous inductive algorithms to interact on a principled, consistent basis and that the inductive inferences being communicated should not be able to introduce unsoundness into the agent community.
منابع مشابه
Towards a Logical Model of Induction from Examples and Communication
This paper focuses on a logical model of induction, and specifically of the common machine learning task of inductive concept learning (ICL). We define an inductive derivation relation, which characterizes which hypothesis can be induced from sets of examples, and show its properties. Moreover, we will also consider the problem of communicating inductive inferences between two agents, which cor...
متن کاملThe Communication of Inductive Inferences
We propose a new approach to communication between agents that perform inductive inference. Consider a community of agents where each agent has a limited view of the overall world. When an agent in this community induces a hypothesis about the world, it necessarily reflects that agent’s partial view of the world. If an agent communicates a hypothesis to another agent, and that hypothesis is in ...
متن کاملTowards a logical model of induction from examples and communication
This paper focuses on a logical model of induction, and specifically of the common machine learning task of inductive concept learning (ICL). We define an inductive derivation relation, which characterizes which hypothesis can be induced from sets of examples, and show its properties. Moreover, we will also consider the problem of communicating inductive inferences between two agents, which cor...
متن کاملTowards a logical model of induction from examples and communication
This paper focuses on a logical model of induction, and specifically of the common machine learning task of inductive concept learning (ICL). We define an inductive derivation relation, which characterizes which hypothesis can be induced from sets of examples, and show its properties. Moreover, we will also consider the problem of communicating inductive inferences between two agents, which cor...
متن کاملPh.D. Thesis Proposal: The Communication of Inductive Inferences
We propose a new approach to communication between agents that perform inductive inference. Previous systems have used methods based on voting strategies or theory refinement to integrate multiple learned models. Our approach uses version spaces as a medium for the inter-agent communication of inductive inferences. This allows multiple models to be integrated using the principle of set intersec...
متن کامل