LCN 03 AI Agents

Intelligent Agent

Anything that can be viewed as perceiving its environment through sensors and acting upon that environment through its effectors to maximise progress towards its goals.

PAGE

• Percepts

• Actions

• Goals

• Environment

Rational Agent

• An agent should strive to do the right thing based on what it can perceive and the actions it can perform.

• The right action will cause the agent to be most successful.

Performance Measure

An objective criterion for success of an agent's behaviour

PEAS

• Performance measure

• Environment

• Actuators

• Sensors

Specifies the settings for intelligent agent designs

Characteristics of Rational Agents:

• Can perform actions in order to modify future percepts so as to obtain useful information

• Autonomous

When is an agent autonomous?

If its behaviour is determined by its own experience (with ability to learn and adapt)

Agent classes

Based on their degree of perceived intelligence and capability.

Rational Action

The action that maximizes the expected value of the performance measure given the percept sequence to date

Russell & Norvig (2003) group agents into five classes based on their degree of perceived intelligence and capability

1. Simple reflex agents

2. Model-based reflex agents

3. Goal-based agents

4. Utility-based agents

5. Learning agents

How many Features we Need to measure for any Agent

1. Autonomy

2. Intelligence

3. Reaction & Pro-action

4. Social Ability

5. Cooperation

6. Migration

Autonomy

How much it is capable of acting on its own.

Intelligence

Ability to learn, solve problems, and adapt to changes.

Level of Reaction

Agents should be both reactive(answers with prompt) and proactive(works without input).

Social Ability

Machine has ability to interact with other systems.

Migration

Portable, and able to move from one system to another.

Environment Types

• Accessible/Inaccessible

• Deterministic/Nondeterministic

• Episodic/Nonepisodic

• Hostile/Friendly

• Static/Dynamic

• Discrete/Continuous

Accessible/Inaccessible

Sensors give access to complete state of the environment.

Deterministic/Nondeterministic

The next state can be determined based on the current state and the action.

Episodic/Nonepisodic

• Episode: each perceive and action pairs

• The quality of action does not depend on the previous episode.

• Sequential

Static/Dynamic

Dynamic if the environment changes during deliberation

Discrete/Continuous

Limited number of actions and each action depends on the other party or can act without waiting.

Semi-Dynamic

If the environment itself does not change with the passage of time but the agent’s performance score does.