compsci - intelligent agents reading

agent

anything that can perceive its environment through sensors and can act upon that environment with effectors

agent = architecture + program

rational agent

the agent that does the action that will cause to be the most successful

performance measure

the criteria that determines how successful an agent is. outside observers establish a standard of what it means to be successful in an environment.

omniscience

an omniscient agent knows the outcomes of its actions and acts accordingly. this concept is impossible to achieve in real life.

rationality

decisions/actions made based on the expected success given what has been perceived by an agent.

doing actions in order to obtain useful information

percept sequence

a complete perceptual history of everything that the agent has perceived so far

ideal rational agent

an ideal rational agent should do whatever action is expected to maximize its performance measure. an agent's behavior depends only on the basis of the evidence provided by the percept sequence and whatever built-in knowledge the agent has

mapping

a list of the actions an agent can make in response to each possible percept sequence

ideal mapping

specifying which action an agent will take in response to any given percept sequence. this provides a design for an ideal agent (see Figure 2.2). the goal is to create agents that can solve a limitless variety of tasks in a limitless variety of environments

autonomy

a system whose behavior is determined by its own experience.

if the agent's actions are solely based on its own built-in knowledge, so much that it does not need to pay attention to its percepts (what it perceives), then it lacks autonomy

agent program

a function that implements the agent mapping from percepts to actions. this is the role of artificial intelligence

architecture

the computing device that the agent program will run on. this makes the percepts from the sensors available to the program, runs the program, and feeds the program's action choices to the effectors

(continue here)

stopped taking notes at page 6 of the pdf, at software agents