Artificial Intelligence: A Modern Approach
| Stuart Russell and Peter Norvig | Bib INPE: 681.3.019 R917A |
Intelligent Agents
An agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through effectors.
Ideal rational agent: For each possible percept sequence, an ideal rational agent should do whatever action is expected to maximize its performance measure, on the basis of the evidence provided by the percept sequence and whatever built-in knowledge the agent has.
Specifying which action an agent ought to take in response to any given percept sequence provides a design for an ideal agent.
A system is autonomous to the extent that its behaviour is determined by its own experience.
Types of agents:
| Simple reflex agents | obey condition-action rules. |
|---|---|
| Agents that keep track of the world | Internal state in order to choose an action. |
| Goal-based agents | Goal information, which describes the desired situations. Search and planning. |
| Utility-based agents | Choose actions based on a definition of utility, a function that maps a state onto a real number. |
In some cases, a utility function can be translated into a set of goals, such that the decisions made by a goal-based agent using those goals are identical to those made by the utility-based agents.