Chapter 13--artificial Intelligence

Question 0

Which kind of ambiguity does this sentence have–lexical, referential, or syntactical? “Go down the street on the left.”

Answer 0

Syntactic– left could refer to the street that is located on your left, or to walk down the left side of the street that you are on.

Question 1

Which kind of ambiguity does this sentence have–lexical, referential, or syntactical? “Stand up for your flag.”

Answer 1

Referential–it isn’t clear who the pronoun “your” refers to.

Question 2

Which kind of ambiguity does this sentence have–lexical, referential, or syntactical? “He drove the car over the lawn mower, but it wasn’t hurt.”

Answer 2

Referential–which did the pronoun “it” refer to, the lawn mower or the car?

Question 3

Which kind of ambiguity does this sentence have–lexical, referential, or syntactical? “I saw the movie flying to Houston.”

Answer 3

Syntactical– which was flying, you or the movie?

Question 4

Which kind of ambiguity does this sentence have–lexical, referential, or syntactical? “Mary and Kay were playing until she came inside.”

Answer 4

Referential– who came inside, Mary or Kay? The pronoun “she” isn’t clear.

Question 5

True/false–a computer does some tasks much better than a human.

Answer 5

True

Question 6

True/false–a human does some tasks much better than a computer.

Answer 6

True

Question 7

True/false–a computer that can pass the Turing test is considered to be intelligent.

Answer 7

True

Question 8

True/false–Some AI researchers don’t think we can achieve true artificial intelligence until a computer processes info in the same way a human mind does.

Answer 8

True

Question 9

True/false–a semantic network is used to model relationships.

Answer 9

True

Question 10

True/false–If info is stored in a semantic network, it is easy to answer questions about it.

Answer 10

False. It depends on the situation

Question 11

True/false–a computer has never beaten a human at master-level chess play.

Answer 11

False.

Question 12

True/false–am inference engine is part of a rule-based expert system.

Answer 12

True

Question 13

True/false–a biological neuron accepts a single input signal and produces multiple output signals.

Answer 13

False, it does the opposite

Question 14

True/false–Each element in an artificial neural network is affected by a numerical weight.

Answer 14

True.

Question 15

True/false–Voice synthesis is the most difficult part of natural language processing.

Answer 15

False

Question 16

True/false–Each human has a unique voiceprint that can be used to train voice recognition systems.

Answer 16

True

Question 17

True/false–the word “light” can be interpreted in many ways by a computer.

Answer 17

True

Question 18

True/false–syntactic ambiguity is no longer a problem for natural language comprehension.

Answer 18

False

Question 19

Fill in the blank– a robot may follow the _______ paradigm to control it’s movements.

Answer 19

Sense-plan-act

Question 20

Who created the 3 fundamental laws of robotics?

Answer 20

Isaac Asimov

Question 21

How is the Turing test organized and administered?

Answer 21

Hugh Loebner started in 1991. Grand prize $100k, responses must be indistinguishable from a human’s.

Question 22

Name and define 2 knowledge- representation techniques.

Answer 22

Semantic network–a knowledge representation technique that focuses on the relationships between objects.

Search tree–a structure that represents all possible moves in a game, for all players in the game.

Question 23

Which object-oriented properties do semantic networks borrow?

Answer 23

Relationships. Like in an entity-relationship diagram. Also inheritance and instantiation.

Question 24

Distinguish between depth-first and breadth-first searching.

Answer 24

Breadth-first checks all options on each level of a search tree before going to the next level. Depth-first checks only one option per level before going deeper to the next level down that specific path.

Question 25

What is meant by pruning a search tree?

Answer 25

Removing options that no human player would consider reasonable.

Question 26

Distinguish between knowledge-based systems and expert systems.

Answer 26

Expert systems are a type of knowledge-based system that uses collected knowledge of human experts to design its if-then rules. Knowledge-based systems are basically rules-based systems that used if-then statements to come up with answers. Broadly-speaking, they are software that uses a specific set of information.

Question 27

Distinguish between rule-based systems and inference engines.

Answer 27

Inference engines are the software that processes if-then queries in a rule-based system.

Question 28

Why is an expert system called a rule-based system?

Answer 28

Because the system works by using rules based around if-then statements which are themselves based on the experts’ collective knowledge.

Question 29

Which part of the software in an expert system determines how the rules are followed and what conclusions can be drawn?

Answer 29

The inference engine

Question 30

How are the rules expressed in an expert system?

Answer 30

As if-then statements

Question 31

What do we call a single cell that conducts a chemically-based electronic signal?

Answer 31

Neuron

Question 32

What do a series of connected neurons form?

Answer 32

A neural network

Question 33

Upon what does the signal along a particular pathway depend?

Answer 33

The chemical composition of the synapse

Question 34

What are the multiple input tentacles in a biological neuron called?

Answer 34

Dendrites

Question 35

What is the output tentacle in a biological neuron called?

Answer 35

Axon

Question 36

From where do dendrites of one neuron pick up the signals from other neurons to form a network?

Answer 36

The synapses

Question 37

What do we call the gap between an axon and a dendrite?

Answer 37

A synapse

Question 38

What tempers the strength of a synapse?

Answer 38

Its chemical composition

Question 39

What is the role of a synapse?

Answer 39

To temper the strength of signals, and carry those same signals from an axon of one neuron to a dendrite of another neuron

Question 40

How is a synapse modeled in an artificial neural network?

Answer 40

By numerical weighting of the signal

Question 41

What is an effective weight in an artificial neuron?

Answer 41

All the inputs multiplied by their effective weights, then added together

Question 42

How is the output value from an artificial neuron calculated?

Answer 42

By checking the effective weight against the threshold value. If the effective weight is lower than the threshold value, a weak signal is output. If the reverse, a strong signal is output.

Question 43

What is a phoneme?

Answer 43

One of the basic sounds that are used in a particular natural language

Question 44

Which issues affect the ability to recognize the words spoken by a human voice?

Answer 44

Accent, speech impediments, mumbling, volume, health of the speaker

Question 45

How can a voice recognition system be trained?

Answer 45

By having a person speak a word several times so the computer can record an average voiceprint from that person. Later speech from that same person can be compared against those voiceprints

Question 46

Name and describe two types of robots.

Answer 46

Fixed robots and mobile ones

Question 47

What are planning systems?

Answer 47

Large software systems that take a given goal, a starting position, and an ending situation, can generate a finite set of actions (plan) that bring about the desired ending situation, if followed.

Question 48

Define “subsumption architecture”.

Answer 48

A set of orders or directions that are prioritized. When orders conflict with each other, then the order with the highest priority is executed.

Question 49

What makes up a robot?

Answer 49

Sensors, actuators, and computational elements.