Overview

CSC 447 - Artificial Intelligence I

What is Artificial Intelligence?

  • Computational models of human behavior?
    (Programs the behave externally like humans)

  • Computational models of human thought?
    (Programs that operate internally the way humans do)

  • Computational systems that behave intelligently?
    (What does it mean to behave intelligently?)

  • Computational systems that behave rationally.

  • AI applications

Agents

  • An agent is software that gathers information about an environment and takes actions based on that information.

  • Examples:

    • a robot

    • a factory

    • a traffic control system

    • ...

Agents and the Environment

  • Formalizing the problem of building an agent. image

World Model

  • \(\mathcal{A}\) - the action space

  • \(\mathcal{P}\) - the percept space

  • \(\mathcal{E}\) - the environment: \(\mathcal{A}* \rightarrow \mathcal{P}\)

  • \(\mathcal{S}\) - the internal state (may not be visible to agent)

  • \(\mathcal{S} \rightarrow \mathcal{P}\) - the perception function

  • \(\mathcal{S} \times \mathcal{A} \rightarrow \mathcal{S}\) - the world dynamics

Agent Design

  • \(\mathcal{U}\) - utility function: \(\mathcal{S} \rightarrow \mathbb{R}\) (or \(\mathcal{S}^* \rightarrow \mathbb{R}\))

  • The agent design problem: find \(\mathcal{P}^* \rightarrow \mathcal{A}\)

  • The agent function maps percept histories to actions.

  • The agent function should maximize the utility of the resulting sequence of states.

Rationality

  • A rational agent chooses whichever action maximizes the expected value of the performance measure given the percept sequence to date.

  • In other words, a rational agent takes actions it believes will achieve its goals

  • Rational \(\neq\) omniscient

    • percepts may not supply all relevant information

  • Rational \(\neq\) clairvoyant

    • action outcomes may not be as expected

  • Hence, rational \(\neq\) successful

  • Rational \(\Rightarrow\) exploration, learning, autonomy

Limited Rationality

  • Problem: the agent may not be able to compute the best action (subject to its beliefs and goals).

  • So, we want to use limited rationality: “acting in the best way you can subject to the computational constraints that you have”

  • The (limited rational) agent design problem: find \(\mathcal{P}^* \rightarrow \mathcal{A}\)

    • mapping of sequences of percepts to actions

    • maximizes the utility of the resulting sequence of states

    • subject to our computational constraints

Thinking

  • Is finding \(\mathcal{P}^* \rightarrow \mathcal{A}\) AI? Aren’t the agents supposed to think?

  • Why is it useful to think? If you were endowed with an optimal table of reactions (\(\mathcal{P}^* \rightarrow \mathcal{A}\)) why do you need to think?

  • Problem: the table is too big; there are too many world states and too many sequences of percepts.

    • In some domains, the required reaction table can be specified compactly.

    • In other domains, we can take advantage of the fact that most things that could happen do not actually happen in practice.

Learning

  • Learning is useful when the agent does not know much about the initial environment or the environment can change.

  • An agent can use sequences of percepts to estimate missing details in the world dynamics.

  • Learning is similar to perception; they both find out about the world based on experience.

    • Perception – short time scale

    • Learning – long time scale

Specifying the Task Environment

  • A task environment is the “problem” that the rational agent is a “solution” for.

  • Specifying a task environment (PEAS)

    • Performance measure

    • Environment

    • Actuators

    • Sensors

Example Task Environment Specification

Taxi driver agent

  • Performance measure? safe, fast, legal, comfortable trip, maximize profits

  • Environment? roads, traffic, pedestrians, customers

  • Actuators? steering, accelerator, brake, signal, horn

  • Sensors? cameras, sonar, GPS, odometer, ...

Properties of Task Environments

  • Fully Observable vs. Partially Observable

    • Can you observe the state of the world directly?

  • Deterministic vs. Stochastic

    • Does an action map one state to a single other state?

  • Episodic vs. Sequential

    • Do current decisions affect future decisions?

  • Static vs. Dynamic

    • Can the world change while you are thinking?

  • Discrete vs. Continuous

    • Are percepts and actions discrete or continuous?

  • Single agent vs. Multiagent

    • Does an agent need to consider the actions of other agents?

Example Task Environments

Solitaire Backgammon Internet shopping Taxi
Observable? Yes Yes No No
Deterministic? Yes No Partly No
Episodic? No No No No
Static? Yes Semi Semi No
Discrete? Yes Yes Yes No
Single-agent? Yes No Yes No

The environment type largely determines the agent design.

Agent Types

  • Four basic agent types in increasing generality.

    • simple reflex agents

    • reflex agents with state

    • goal-based agents

    • utility-based agents

  • All the basic agent types can be turned into learning agents.

Simple Reflex Agents

image

Reflex Agents with State

image

Goal-Based Agents

image

Utility-Based Agents

image

Learning Agents

image