Intelligent Agents and Rationality: Key Concepts and Environments

1. Plans for Today • Chapter 2: Intelligent Agents (until break) • Lisp: Some questions that came up in lab • Resume intelligent agents after Lisp issues

2. Intelligent Agents • Agent: anything that can be viewed as… • perceiving its environment through sensors • acting upon its environment through actuators • Examples: • Human • Web search agent • Chess player • What are sensors and effectors for each of these?

3. Rational Agents • Conceptually: one that does the right thing • Criteria: Performance measure • Performance measures for • Web search engine? • Tic-tac-toe player? Chess player? • When performance is measured plays a role • short vs. long term

4. Rational Agents • Omniscient agent • Knows actual outcome of its actions • What info would chess player need to be omniscient? • Omniscience is (generally) impossible • Rational agent should do right thing based on knowledge it has

5. Rational Agents • What is rational depends on four things: • Performance measure • Percept sequence: everything agent has seen so far • Knowledge agent has about environment • Actions agent is capable of performing • Rational Agent definition: • Does whatever action is expected to maximize its performance measure, based on percept sequence and built-in knowledge

6. Autonomy • “Independence” • A system is autonomous if its behavior is determined by its percepts • An alarm that goes off at a prespecified time is not autonomous • An alarm that goes off when smoke is sensed is autonomous • A system without autonomy lacks flexibility

7. The Task Environment • An agent’s rationality depends on • Performance Measure • Environment • Actuators • Sensors What are each of these for: • Chess Player? • Web Search Tool? • Matchmaker? • Musical performer?

8. Environments: Fully Observable vs. Partially Observable • Fully observable: agent’s sensors detect all aspects of environment relevant to deciding action • Examples? • Which is more desirable?

9. Environments: Determinstic vs. Stochastic • Deterministic: next state of environment is completely determined by current state and agent actions • Stochastic: uncertainty as to next state • If environment is partially observable but deterministic, may appear stochastic • If environment is determinstic except for actions of other agents, called strategic • Agent’s point of view is the important one • Examples? • Which is more desirable?

10. Environments: Episodic vs. Sequential • Episodic: Experience is divided into “episodes” of agent perceiving then acting. Action taken in one episode does not affect next one at all. • Sequential typically means need to do lookahead • Examples? • Which is more desirable?

11. Environments: Static vs. Dynamic • Dynamic: Environment can change while agent is thinking • Static: Environment does not change while agent thinks • Semidynamic: Environment does not change with time, but performance score does • Examples? • Which is more desirable?

12. Environments: Discrete vs. Continuous • Discrete: Percepts and actions are distinct, clearly defined, and often limited in number • Examples? • Which is more desirable?

13. Environments: Single agent vs. multiagent • What is distinction between environment and another agent? • for something to be another agent, maximize a performance measure depending on your behavior • Examples?

14. Structure of Intelligent Agents • What does an agent program look like? • Some extra Lisp: Persistence of state (static variables) • Allows a function to keep track of a variable over repeated calls. • Put functions inside a let block • (let ((sum 0)) (defun myfun (x) (setf sum (+ sum x))) (defun report () sum))

15. Generic Lisp Code for an Agent • (let ((memory nil)) (defun skeleton-agent (percept) (setf memory (update-memory memory percept)) (setf action (choose-best-action memory)) (setf memory (update-memory memory action)) action ; return action ))

16. Table Lookup Agent • In theory, can build a table mapping percept sequence to action • Inputs: percept • Outputs: action • Static Variable: percepts, table

17. Lookup Table Agent • (let ((percepts nil) (table ????) (defun table-lookup-agent (percept) (setf percepts (append (list percept) percepts)) (lookup percepts table)) ))

18. Specific Agent Example:Pathfinder (Mars Explorer) • Performance Measure: • Environment: • Actuators: • Sensors: • Would table-driven work?

19. Four kinds of better agent programs • Simple reflex agents • Model-based reflex agents • Goal-based agents • Utility-based agents

20. Simple reflex agents • Specific response to percepts, i.e. condition-action rule • if new-boulder-in-sight thenmove-towards-new-boulder • Advantages: • Disadvantages:

21. Model-based reflex agents • Maintain an internal state which is adjusted by each percept • Internal state: looking for a new boulder, or rolling towards one • Affects how Pathfinder will react when seeing a new boulder • Can be used to handle partial observability by use of a model about the world • Rule for action depends on both state and percept • Different from reflex, which only depends on percept

22. Goal-Based Agents • Agent continues to receive percepts and maintain state • Agent also has a goal • Makes decisions based on achieving goal • Example • Pathfinder goal: reach a boulder • If pathfinder trips or gets stuck, can make decisions to reach goal

23. Utility-Based Agents • Goals are not enough – need to know value of goal • Is this a minor accomplishment, or a major one? • Affects decision making – will take greater risks for more major goals • Utility: numerical measurement of importance of a goal • A utility-based agent will attempt to make the appropriate tradeoff

24. Lisp Questions

25. Why the dot in cons?Two Explanations: • High level: cons expects a list in the second position • Lower level: • Cons takes a cons cell from the free storage list • Puts first argument in “first” position • Puts second argument in “rest” position • Separates by a dot, unless “rest” position is a pointer (indicates continuing list)

26. How does append work? • Makes copy of first list • Takes last pointer and points to second list • Picture

27. How to debug? • Can trace function calls with (trace function) and (untrace function) • Demonstration with mystery function from lab • At a Break> prompt, can see call stack with • backtrace • Can go through code step by step • (step (mystery 2 3)) • Use step and next to go through each function as you go along • Use (print var)

28. Random bits Why the p in (zerop x)? • p = predicate • NOT true that p = positive

29. Scoping and binding • let declares a scope where variable bindings are insulated from outside • usual notions of local and global variables apply • if you want to change a global variable from within a function