Lecture on Rational Agents

Overview -> Rational Agents -> lecture


  • What is AI?
    • Act or think? Humanly or rationally?

The Agent

Agent is the perceive-think-act cycle
Agent is the perceive-think-act cycle
  • Agent: perceive and act
    • vs machine learning
  • A rational agent is one that chooses the right action.
    • what is right?
    • Hume on is-ought
  • Predefined goal
  • Aristotle’s algorithm implemented by Newell & Simon (GPL)
    • means-ends analysis
    • we deliberate on means, not on ends


  • PEAS - Performace Measure, Environment, Actuators, Sensors

Task Environment

  • Properties of the Environment
    • Fully or partially observable
      • can you sense all relevant aspects?
    • Single or multi-agent
      • what is an agent?
      • do actions influence other agent?
    • game theory - can you predict other agent’s actions?
    • Deterministic or not
    • Episodic or sequential
      • memory
    • Static or dynamic
      • can it change while you think?
    • round-based games are static
    • Discrete or continuous
      • discrete state space
    • discrete time
    • Known or unknown
      • do you have a complete model of the world?
    • a known world may have unknown or stochastic states

Paradigms or Program Type

  1. simple reflex agent
  2. model-based agents
  3. goal-based agents
  4. utility-based agents
  5. learning agents
  • Problem Generator