WEEK ONE

Question 1

3 examples of AI approaches

Answer 1

Physical Symbol System (Symbolic) 
Neural Network (Sub-Symbolic). Not a symbol 
Embodied Cognition (Sub-symbolic)

Question 2

Who are the ‘god fathers of AI’?

Answer 2

Herb Simon and Allen Newell: Physical Symbol System Hypothesis

Question 3

What is a symbol?

Answer 3

Meaningful pattern that can be manipulated. Language Like representation.
A symbolic representation

Question 4

What is problem solving/ problem space?

Answer 4

Problem solving = searching through the problem space until a solution state is found
Problem Space
- Set of all possible problem ‘states’
- Includes states and operators
- can be a tree/graph but two nodes can only be connected by one path

Question 5

What are the 3 states?

Answer 5

State: Representation of the problem 
Initial State
Goal State 
Intermediate State: state on the way to the goal 
These are all represented by nodes

Question 6

What does the path from an initial state to a goal state represent?

Answer 6

A solution

Question 7

How to get from an initial state to a goal state

Answer 7

Search Strategy: Which states are visited (or generated first)
There are 4 ways

Question 8

Two types of AI search algorithms

Answer 8

Uninformed Search: No domain specific knowledge used

Informed Search:Heuristics are used

Question 9

What is an algorithm?

Answer 9

Recipe that specifies how a task can be accomplished

Question 10

Breadth First Search (BFS)

Answer 10

Generates or visits all the does in a given solution tree before continuing to the next level
1) Add the root node
2) Test to see if the first element i queue is the goal node.
3) Remove the first element from the front and add that
elements children to the back.
Does not use heuristics

Question 11

Depth First Search (DFS)

Answer 11

Expands the left most children down all the levels. Going down.
Remove the front and add to the front
Stack
Does not use heuristics

Question 12

Heuristic

Answer 12

“Estimate of how many operations we need to apply to get from the current state to the goal state”

Informed Search. Gives agent an estimate of how far it is from the goal. Needs to estimate a path cost from the current state to the goal state.

Question 13

Hill Climbing

Answer 13

Variant of DFS
Remove the top element from the stack, add the elements children to the top of the stack in order (lowest cost first)
Ex. S
A4 B1 –> B1 A4

Question 14

Best First Search Strategy

Answer 14

Selects all nodes with best estimated cost with all nodes considered in a given step

Question 15

Psychical Symbol System

Answer 15

“necessary and sufficient means for general intelligent action

Intelligent Action: Representations are symbolic as well as the transformation of rules or operations