Term
|
Definition
The way the system is, at a moment in time.
nothing about possible changes |
|
|
Term
|
Definition
Determines if a given state satisfies the goal of the problem |
|
|
Term
|
Definition
A path from inital state to a state that satifies the goal state |
|
|
Term
|
Definition
In optimal solution is a solution with least cost |
|
|
Term
|
Definition
- Inital state (definition)
- a set of operators/ actions to go from one state to another
- states and operators define the state space
- goal test
- path and funtions
- solution |
|
|
Term
|
Definition
- no knowledge about the distance (cost) from the current state to the goal state
- only detect when a goal state has been found
(brute force/blind search)
|
|
|
Term
|
Definition
- access to extra information (an estimation of cost for the remaining path from current to goal state
- estimation by heuristic function
- "informed search" |
|
|
Term
|
Definition
- State in the state space to which the node corresponds |
|
|
Term
|
Definition
- node which was expanded to generate this node |
|
|
Term
|
Definition
- Operator used to generate this node |
|
|
Term
|
Definition
distance from root node to this one |
|
|
Term
|
Definition
Cost of getting from the root node to this one |
|
|
Term
|
Definition
The programmer specifies a goal to be achieved and the system works out how to achieve it
e.g. prolog |
|
|
Term
|
Definition
The programmer specifies how to achieve a goal
e.g. C, C++, Java |
|
|
Term
Advantages of Logic Programming |
|
Definition
- Ease of programming (prolog programs are smaller in size relative to procedual)
- Increases programmer productivity
- Facilitates proving correctness of programs |
|
|
Term
Disadvantages of logic programming |
|
Definition
- Slows the execution of programs |
|
|
Term
|
Definition
member(X,[X|T]).
member(X,[H|T]) :- member(X,T). |
|
|
Term
Paper Needed
Take a list and give a copy of it |
|
Definition
clist([],[]).
clist([H|T],[H|R]) :- clist(T,R). |
|
|
Term
Paper Needed
last element |
|
Definition
last(X,[X]).
last(X, [H|T]):-last(X,T). |
|
|
Term
Length of a list predicate |
|
Definition
len([], 0).
len([H|T],N):- len(T,X), N is X+1. |
|
|
Term
Check for integars predicate |
|
Definition
intlist([]).
intlist([X]) :- integer(X).
intlist([H|T]):- integer(H), intlist(T). |
|
|
Term
|
Definition
max2(X,Y,Z):- X>=Y, Z is X.
max2(X,Y,Z):- X < Y, Z is Y.
maxlist([X],X).
maxlist([H|T], M):- maxlist(T, Mt), max2(H,Mt,M). |
|
|
Term
Postive elements in a list |
|
Definition
io([]).
io([H|T]):- H>0, write (H), io(T).
io([H|T]):- H <= 0, io(T). |
|
|
Term
|
Definition
del(H, [H|T], T).
del(X,[H|T], [H|T1]):- del(X, T, T1). |
|
|
Term
|
Definition
evenlength([]).
evenlength([H|T]):- oddlength(T).
oddlength([H|T]) :- evenlength(T). |
|
|
Term
Write a Prolog predicate ‘mul’ that takes a list of numbers and multiplies each element of the list by 2. |
|
Definition
mul([],[]).
mul([H|T], Result) :- mul(T,R) x is Z * H, concat([X], R, Result). |
|
|
Term
Write a Prolog predicate ‘add’ that takes a term ‘X’ and a list L and adds X to L as the first element if X is not already in L. |
|
Definition
member(X, [X | T]).
member(X, [H | T]) :- member(X, T).
add(Element, List, List) :- member(Element, List), !. add(Element, List, [Element | List]). |
|
|
Term
|
Definition
run:- read(X), test(X).
test(1) :- !, write('Invalid Number').
test(_) :- write('Invalid Input'). |
|
|
Term
Classify into 3 classes: positive, zero and negative |
|
Definition
classify(Number, positive):- Number > 0, !.
classify(0, zero).
classify(Number, negative) :- Number < 0, !. |
|
|
Term
|
Definition
factorial(0,1).
factorial(N, Result):-
N > 1,
N1 is N-1,
factorial(N1, Result),
Result is Result1 * N. |
|
|
Term
|
Definition
- Expand the shallowest nodes on the fringe first
- All nodes at depth d are expanded before any nodes at depth d + 1 |
|
|
Term
|
Definition
Time Complexity: bd
Space Complexity: bd
Optimal: yes
Completeness: yes |
|
|
Term
Uniform Cost Search Criteria |
|
Definition
Time Complexity: bd
Space Complexity: bd
Optimal: yes
Completeness: yes |
|
|
Term
|
Definition
Time Complexity: bm
Space Complexity: bm
Optimal: no
Completeness: no |
|
|
Term
|
Definition
Time Complexity: bl
Space Complexity: bl
Optimal: no
Completeness: yes if l > d |
|
|
Term
Iterative Deepening Search Criteria |
|
Definition
Time Complexity: bd
Space Complexity: bd
Optimal: yes
Completeness: yes |
|
|
Term
|
Definition
Time Complexity: bd/2
Space Complexity: bd/2
Optimal: yes
Completeness: yes |
|
|
Term
|
Definition
The data from which a patten must be discovered |
|
|
Term
|
Definition
The actual funciton that describes the data. This is unknown |
|
|
Term
|
Definition
any function that can be induced from the training data is called hypothesis |
|
|
Term
|
Definition
- close to the target function. I.e. minimises error
- describes a large number of instances in a concise way
- generalises well to unseen test data
|
|
|
Term
If we have more than one hypothesis we use.. |
|
Definition
Ockhams Razor Princiable
- This is the most likely hypothesis is the simpliest one that is consistent with the training data |
|
|
Term
|
Definition
A concept is a boolean- valued function defined over some set (binary classification) |
|
|
Term
|
Definition
- '?' which indiciates that any value is acceptable for the attribute
- specify a single value for the attribute
- Ø indicates that no value is acceptable for the attribute |
|
|
Term
|
Definition
- Guarenteed to output the most specific hypothesis within H that is consistent with the positive training example
- Output of FIND-S will be consistent with the negative training examples
- Finds only one of the possible hypothesis |
|
|
Term
LIST-THEN-ELIMINATE Pros and Cons |
|
Definition
- Requires H to be finite
- Guarenteed to output all hypothesis consisten with the training data
- Requires exhaustive ennumeration of hypothesis in H |
|
|
Term
Concept Learning Methods Pros and Cons |
|
Definition
- Methods are conceptually simple
- Can be used to represent attribute value pairs
- Difficult to use these method with structured data |
|
|
Term
Parts of the decision tree |
|
Definition
- Each non-leaf node specifies an attribute
- Each leaf node has an integer value
- Each edge from a non leaf node is labled with a value for the attribute specified by the node
|
|
|
Term
|
Definition
- Calculate the info gain for all attributes
- The attribute with the highest info gain in the root node
- At every level, choose the attribute with the highest info gain |
|
|
Term
|
Definition
- only use a part of the training data for leaving the hypothesis
- use the remaining training data for testing the hypothesis
- repeat this with different subsets of data, and choose a hypothesis with good prediction performance |
|
|
Term
|
Definition
- There is noise if 2 or more examples have the same description in terms of attributes but different classifications |
|
|
Term
|
Definition
- Take the majority classifcation for the set examples
- consider the estimated probablities for each classification using their relative frequency
|
|
|
Term
|
Definition
- Hypothesis is said to overfit if it imposes meaningless regularity on it bu including irrelevant attributes in descion making |
|
|
Term
|
Definition
- Prune irrelevant attributes
- cross validate hypothesis and choose best at decsion making |
|
|
Term
Logic based learning (resolution) |
|
Definition
If a hypothesis and an example are inconsistent, a logical inference system can learn from the example by eliminating the hypothesis. |
|
|
Term
What is current best hypothesis |
|
Definition
i. Start with an initial hypothesis and call it the current hypothesis
ii. For each training instance check if it is consistent with current hypothesis and if not adjust the current hypothesis to make it consistent |
|
|
Term
Two ways to make it consistent (current best hypothesis) |
|
Definition
1. Generalisation – dropping conditions
2. Specialised – Adding conditions |
|
|
Term
Pros and Cons of Current Best Hypothesis |
|
Definition
1. This method is conceptually simple
2. Non-deterministic; so several generalisations and specialisations can be made
3. Choices made may not lead to simple hypothesis
4. Choices may lead to irrecoverable situations, so you must backtrack |
|
|
Term
What is linear separate data |
|
Definition
i. Data is linearly separable if a straight line can separate positive and negative instances |
|
|
Term
What is Perceptron Learning
|
|
Definition
i. For the classification of linearly separable data using a binary classifier, a perceptron, It receives info from perceptrons/environment, processes it then transmits to other perceptrons/environments |
|
|
Term
Pros and Cons of Perceptron Learning |
|
Definition
1. For linearly separable training data, guaranteed to find a solution in a finite number of steps
2. Conceptually simple
3. Multiple solutions possible
4. May not converge for training data that isn’t linearly separable
5. Does not readily generalise more than 2 classes |
|
|
Term
|
Definition
a. Used to classify data that is
i. Not linearly separable
ii. Represented in Euclidean space
b.
Do not construct hypothesis
c.
Each time new data is encountered a prediction is made based on new to old |
|
|
Term
|
Definition
This method is for classification of data represented in Euclidean space. It is used to classify data that is not linearly separable. In order to classify a point xq:
• find its k nearest neighbors by using Euclidean distance measure
• then, find the majority class for these k neighbors and classify xq as the majority class. |
|
|
Term
Pros and Cons of K neighbour |
|
Definition
i. Pros
1. Useful local approximation of complex non-linearly separable training data
ii. Cons
1. The cost of classifying new test data can be high since computation must be repeated for each test data
2. The classification of test data os sensitive to the value of k |
|
|
Term
Pros of Uniformed Searches |
|
Definition
i. Advantages
1. Useful for graph search or traversal
2. BFS – complete, simple to implement on any search problem
3. DFS – memory O(bm)
4. Iterative – memory O(bm)
5. Uniform – optimal |
|
|
Term
|
Definition
a. Uninformed Search
i. Disadvantages
1. Has no knowledge about distance from the current state to the goal
2. Can only detect when a goal state has been found
3. BFS – Memory is high O(b^d) as its inefficient
4. DFS – not complete
5. Iterative – not optimal
6. Uniform – exponential memory |
|
|
Term
Advantages pf Informed Search |
|
Definition
1. A* -If h is admissible and consistent then is optimal and complete
2. Good to have an estimate
3. Can solve problems quicker |
|
|
Term
Disadvantages of Informed |
|
Definition
i. Disadvantages
1. Need to keep everything that happens in the memory
2. Need to be able to do quick cost comparisons
3. Need to choose good heuristics
4. Greedy – Not complete or optimal, can be false starts
5. A* - high memory required |
|
|
Term
|
Definition
a. A sentence S is satisfiable if there exists a structure that S evaluates to true. Otherwise S is unsatisfiable or contradictory
b. A sentence S is valid or a tautology if for each possible structure for S, S evaluates to true |
|
|
Term
Propostitonal Logic Pros and Cons |
|
Definition
i. Hard to express properties of objects or relations between them
ii. Propositional logic ant deal with facts changing over time
iii. Propositional logic is sound, complete, monotonic and consistent
iv. Propositional logic is decidable |
|
|
Term
First Order Logic Pros and Cons |
|
Definition
i. Assumes the world contains objects and relations or properties with them
ii. Contains quantifiers
iii. More expressive |
|
|
Term
|
Definition
- Hill climbing starts by choosing a random start point and setting it to current point - It then searches the neighbourhood(comprised of point n1,...nm) of c - It then chooses the best among (n1,...nm) to be the current point - the above steps are then repeated until no improvement can be made to the current point |
|
|
Term
|
Definition
A local optimum is a peak in a landscape, it is optimum around it's immediate surrounding neighbour hood |
|
|
Term
|
Definition
A global optimum is an optimum over the entire landscape. |
|
|
Term
|
Definition
This is process of taking two atomic sentences and giving a substitution that makes the two sentences look the same, or fails if there is no such subsittion.
A most general unifier is a substitution which makes the least commitment about bindings of the variables |
|
|
Term
|
Definition
This is the process of eliminating existential quantifies and replacing existentially quantified variables with Skolem functions |
|
|
Term
|
Definition
A program is said to learn from experience E with respect to some class of tasks in T, as measure by P, improves by performance |
|
|
Term
Advantages of Perception learning |
|
Definition
for linearly separable training data, guaranteed to find a solution in a finite number of steps
- it is conceptually simple |
|
|
Term
Disadvantages of Perception learning |
|
Definition
- Multiple solutios possible - May not converge for training data that is not linearly separable - Does not readily generalise to more than two classes |
|
|
Term
|
Definition
And Elimination And Introduction Or Introduction Resolution Unit resolution Mod Pod Double negation
Contrapositive Rule: (P ⇒ Q) ≡ (¬Q ⇒ ¬P)
de Morgan’s Laws: ¬(P ∧ Q) ≡ (¬P ∨ ¬Q) and ¬(P ∨ Q) ≡ (¬P ∧ ¬Q)
Implication Elimination: (P ⇒ Q) ≡ (¬P ∨ Q) |
|
|
Term
|
Definition
concat ([ ], L, L). concat([ H | T ], L2, [H | L3]):- concat(T, L2, L3). |
|
|