AI: Artificial Intelligence Lecture 3 AI: Artificial Intelligence

Agents An agent is anything that can be viewed as perceiving (getting input) its environment through sensors and acting upon that environment through actuators (tools used for actions).

Human Vs Robotic Vs Software Agent A human agent has eyes, ears, and other organs for sensors and hands, legs, vocal tract, and so on for actuators. A robotic agent might have cameras and infrared range finders for sensors and various motors for actuators. A software agent receives keystrokes, file contents, and network packets as sensory inputs and acts on the environment by displaying on the screen, writing files, and sending network packets.

Perception A  percept  is the input that an intelligent agent is perceiving at any given moment. An percept sequence agent’s percept sequence is the complete history of everything the agent has ever perceived. In general, an agent’s choice of action at any given instant can depend on the entire percept sequence observed to date, but not on anything it hasn’t perceived.

Example: Vacuum-Agent Percepts: Location and status, e.g., [A,Dirty] Actions: Left, Right, Suck, NOp function Vacuum-Agent([location,status]) returns an action if status = Dirty then return Suck else if location = A then return Right else if location = B then return Left A percept is the input that an intelligent agent  is perceiving at any given moment.

Rational agents For each possible percept sequence, a rational agent should select an action that is expected to maximize its performance measure, given the evidence provided by the percept sequence and the agent’s built-in knowledge Performance measure (utility function): An objective criterion for success of an agent's behavior. (devised on environment states, not on agents states). ExpectedUtility(action) = sum_outcomes Utility(outcome) * P(outcome|action)

Specifying the task environment Problem specification: Performance measure, Environment, Actuators, Sensors (PEAS) Example: Autonomous Taxi Performance measure Safe, fast, legal, comfortable trip, maximize profits Environment Roads, other traffic, pedestrians, customers Actuators Steering wheel, accelerator, brake, signal, horn Sensors Cameras, speedometer, GPS, engine sensors, keyboard

Another PEAS example: Spam filter Performance measure Minimizing false positives, false negatives Environment A user’s email account, email server Actuators Mark as spam, delete, etc. Sensors Incoming messages, other information about user’s account

Medical Diagnosis System Performance measure Healthy Patients, Minimize cost Environment Patients, Hospital, Staff Actuators Skin Display (questions, tests, diagnosis, treatments) Sensors Keyboard(entry of symptoms, patients answers)

Environment types Fully observable vs. partially observable Deterministic vs. stochastic Episodic vs. sequential Static vs. dynamic Discrete vs. continuous Single agent vs. multi-agent Known vs. unknown A stochastic process is simply a collection of random variables indexed by time. It will be useful to consider separately the cases of discrete time and continuous time. 

Fully observable vs. partially observable Do the agent's sensors give it access to the complete state of the environment? vs.

Fully Observable (accessible) If an agent’s sensors give it access to the complete state of the environment at each point in time, then we say that the task environment is fully observable. the sensors detect all aspects that are relevant to the choice of action; relevance, in turn, depends on the performance measure. Fully observable environments are convenient because the agent need not maintain any internal state to keep track of the world.

Partially Observable (inaccessible) An environment might be partially observable because of noisy and inaccurate sensors or because parts of the state are simply missing from the sensor data Examples: chess – the board is fully observable Autonomous Taxi – what is around the next bend is not observable (yet).

Deterministic vs. stochastic Is the next state of the environment completely determined by the current state and the agent’s action? Strategic: the environment is deterministic except for the actions of other agents vs.

Deterministic Vs Stochastic Deterministic the next state of the environment is completely predictable from the current state and the action executed by the agent Stochastic the next state has some uncertainty associated with it, Uncertainty could come from randomness, lack of a good environment model, or lack of complete sensor coverage Stochastic= variable (temperature) Deterministic= constant (time table of an inst)

Episodic vs. sequential Is the agent’s experience divided into unconnected episodes, or is it a coherent sequence of observations and actions? vs. Episodic: The agent's experience is divided into atomic “episodes,” and the choice of action in each episode depends only on the episode itself

Episodic Vs Sequential Episodic The agent's experience is divided into atomic "episodes" (each episode consists of the agent perceiving and then performing a single action) Sequential if current decisions affect future decisions, or rely on previous ones Examples of episodic are expert advice systems – an episode is a single question and answer Packman is an example of Sequential Many are both – a number of episodes containing a number of sequential steps to a conclusion

Static vs. dynamic Is the world changing while the agent is thinking? Semidynamic: the environment does not change with the passage of time, but the agent's performance score does vs.

Static Vs Dynamic Chess with clock is semidynamic Static if nothing (other than the agent) in the environment changes Dynamic if the environment may change over time. If the environment itself does not change with the passage of time but the agent’s performance score does, then we say the environment is semidynamic. CrossWord Puzzles are Static, aging, integers Taxi Driving is dynamic Chess with clock is semidynamic

Discrete vs. continuous Does the environment provide a fixed number of distinct percepts, actions, and environment states? Time can also evolve in a discrete or continuous fashion vs.

Discrete vs. continuous Discrete time moves in fixed steps, usually with one measurement per step (and perhaps one action, but could be no action). E.g. a game of chess Continuous, Signals constantly coming into sensors, actions continually changing. E.g. driving a car

Single-agent vs. multi-agent Is an agent operating by itself in the environment? vs.

Single Vs Multi Agent An agent operating by itself in an environment is single agent. Multi agent is when other agents are present A strict definition of an other agent is anything that changes from step to step, and can sense and act. Other players in a football team (or opposing team), wind and waves in a sailing agent, other cars in a taxi driver

Known vs. unknown Are the rules of the environment (transitions and rewards) known to the agent? Strictly speaking, not a property of the environment, but of the agent’s state of knowledge vs.

known environment, the outcomes (or outcome probabilities if the environment is stochastic) for all actions are given. if the environment is unknown, the agent will have to learn how it works in order to make good decisions. The distinction between known and unknown environments is not the same as the one between fully and partially observable environments. It is quite possible for a known environment to be partially observable—for example, in solitaire card games, I know the rules but am still unable to see the cards that have not yet been turned over. Conversely, an unknown environment can be fully observable—in a new video game, the screen may show the entire game state but I still don’t know what the buttons do until I try them.

Examples of different environments Word jumble solver Chess with a clock Scrabble Autonomous driving Fully Fully Partially Partially Observable Deterministic Episodic Static Discrete Single agent Deterministic Strategic Stochastic Stochastic Episodic Sequential Sequential Sequential Static Semidynamic Static Dynamic Discrete Discrete Discrete Continuous Single Multi Multi Multi