U18 Artificial Intelligence / U19

Question 1

artificial intelligence

Answer 1

the ability of a computer to carry out tasks that are usually done by humans because they require human intelligence

Question 2

A* algorithm

Answer 2

adds an extra heuristic value (smart guess) on how far we have to go to reach the destination most efficiently

Question 3

machine learning

Answer 3

a subset of AI in which algorithms are trained and learn from their past experience

Question 4

deep learning

Answer 4

• subset of machine learning inspired by the structure of the human brain called artificial neural networks
• extract pattern from data using neural networks

Question 5

unlabeled data

Answer 5

data to which a label is not attached

Question 6

labelled data

Answer 6

unlabeled data becomes labelled the moment a meaning is attached (label)

Question 7

supervised learning

Answer 7

using known tasks with given outcomes to enable a computer program to improve its performance in accomplishing similar tasks

Question 8

unsupervised learning

Answer 8

using a large number of tasks with unknown outcomes to enable a computer program to improve its performance in accomplishing similar tasks

Question 9

reinforcement learning

Answer 9

using a large number of tasks with unknown outcomes and the use of feedback to enable a computer program to improve its performance in accomplishing a similar task

Question 10

artificial neural networks

Answer 10

• based on the interconnection between neurons in the human brain. the system is able to think like a human using these neural networks and its performance improves with more data
• the hidden layers are where data from the input layer is processed into something that can be sent to the output layer

Question 11

why are there multiple hidden layers in an ANN?

Answer 11

• enables deep learning to take place
• where the problems you’re trying to solve has a higher level of complexity it requires more layers to solve
• to enable neural network to learn and make decisions on its own
• to improve the accuracy of results

Question 12

back propagation

Answer 12

• initial outputs from the system are compared to the expected output
• the system weightings are adjusted to minimise the difference between actual and expected result
• once the errors in the output have been eliminated, the neural network is functioning correctly
• if the errors are still too large, the weightings are altered

Question 13

machine learning vs deep learning

Answer 13

• ML enables machines to make decisions on their own based on past data while DL enables machines to make decisions using an artificial neural network
• ML needs only a small amount of data to carry out the training while DL requires large amounts of data for the training stage
• for ML, most features in the data need to be identified in advance and then manually coded into the system while DL learns the features of the data from the data itself and doesn’t need to be identified in advance

Question 14

regression

Answer 14

-a way of analysing data before it is input into a system
- used to make predictions from given data by leaning some relationship between input and output

Question 15

how do artificial neural networks enable machine learning

Answer 15

• weights are assigned for each connection b/w nodes in the ANN
• the data is input within the input layer and is passed into the system
• data is analysed at each subsequent hidden layer where outputs are calculated
• this process of learning is repeated many times to achieve optimum outputs
• the deep learning network will have created complex feature detectors
• output layer provides the results
• back propagation of errors will be used to correct any errors that have been made

Question 16

big O notation

Answer 16

mathematical notation used to describe the performance or complexity of an algorithm in relation to the time taken or the memory used for the task

Question 17

O(1)

Answer 17

describes an algorithm that always takes the same time to perform the task. e.g: an algorithm deciding if a number is positive or negative

Question 18

O(N)

Answer 18

describes an algorithm where the time increases linearly in relation to the number of items (N). e.g: linear search

Question 19

O(N^2)

Answer 19

describes an algorithm where the time to perform the task will increase linearly to the square of the number of items (n). e.g: bubble sort, insertion sort

Question 20

O(2^N)

Answer 20

describes an algorithm where the time to perform the tasks double every time the algorithm uses an extra item of data

Question 21

O(logN)

Answer 21

describes san algorithm where the time increases logarithmically. e.g: binary search

Question 22

stage 1: lexical analysis

Answer 22

• white spaces/redundant characters/comments are removed
• identification of errors
• converts a sequence of characters into a sequence of tokens (tokenisation)
  -using a keyword table that contains all the tokens for the reserved words and symbols (contains reserved keywords and operators used)

Question 23

stage 2: syntax analysis

Answer 23

• checks that the rules of grammar have been obeyed
• produces a parse tree to interpret the meaning of a sequence of tokens
• produces an error report

Question 24

stage 3: code generation

Answer 24

• converts an intermediate representation of source code into an executable form
• source code must be syntactically correct for successful generation of an object program

Question 25

stage 4: code optimistation

Answer 25

• minimises resources used (storage space, memory, CPU)
• remove redundant code as fewer instructions are required
• code reorganised to make it more efficient
• used to minimise execution time