General Flashcards
What is the universal definition for AI
There isn’t one, but there are some ideas:
- systems that act like humans
- Systems that think like humans
- Systems that act rationally
- Systems that think rationally
What does it mean to act rationally?
Doing what is expected to maximize goal achievement
What does a rational agent do?
Acts to achieve best outcome, or best expected outcome from the given information.
What is the turing test?
A test for checking wherher a computer acts human-like.
An interogator will ask some questions and if the interogator doesn’t know if it is talking to a computer, it succeeds
What is cognitive science
Study of the mind and its processes.
I.e. How humans think, learn, perceive, remember and interact with the world.
What is an agent function?
A function that maps perceptions to an action:
f: P* -> A
What foundation is symbolic AI built upon?
Logic, rules and symbolic representations
What are the strengths and weaknesses of symbolic AI?
Strengths: Clear and explicit reasoning. Excels in structured tasks
Weaknesses: Struggles with perception
What is neuro AI
A type of AI that uses the concept of biological neural networks
What are the strengths and weaknesses of neuro AI
Strengths: Excels in perception tasks, adats to unstructured data
Weaknesses: needs large data and high computational power
What is AGI?
Stands for “Artificial General Intelligence”
An AGI is capable of understanding, learning, and applying knowledge across a wide range of tasks at a level comparable to that of a human.
What is the difference between weak and strong AI
Weak AI: Act as if it is intelligent.
Strong AI: Actually consciously thinking.
What is a performance measure?
Something that evaluates a sequence of environment states, to check whether it is desireable.
What four things does the rationality of an agent depend on?
- The performance measure
- The agent’s prior knowledge of the environment
- The actions the agent can perform
- The agent’s percept sequence to date
What is meant by a percept sequence?
The complete history of everything the agent has ever perceived.
What is meant by autonomy for a rational agent?
That the agent learns to compensate for partial or incorrect prior knowledge.
What does PEAS stand for?
P: Performance (measure)
E: Environment
A: Actuators
S: Sensors
Name different types of properties of task environments
Fully/partially observable
Single-agent/multi-agent
Deterministic/nondeterministic/stochastic
Episodic/sequential
Static/dynamic
Discrete/Continous
Known/unknown
What is the difference between deterministic and nondeterministic environments?
Deterministic: You know what state you will end up in after taking an action
Non-deterministic: You don’t know what state you will end up in after taking an action.
What is the difference between episodic and sequential environments?
Episodic: Agent receives a percept and performs a single action. Next episode does not depend on actions taken in prior episodes.
Sequential: Current decision can have effect on later situations.
What is the difference between static and dynamic environments?
Dynamic: Environment can change while an agent is processing a decision.
Static: Environment does not change while an agent is processing a decision.
What is the difference between discrete and continuous environments?
Discrete: Number of reachable states from a state is countable.
Continuous: Infinite number of possible actions.
What is the difference between known and unknown environments?
Known: All possible outcomes for all actions are given.
Unknown: Not all possible outcomes for all actions are given.
Can a known environment be partially observable?
Yes. Example: Solitaire. You know all rules, but are not able to see the cards that have not been turned over.
