Introduction To Artificial Intelligence: Amna Iftikhar Spring ' 2021 1

Introduction to Artificial
Intelligence
Lecture 05
Amna Iftikhar Spring ' 2021 1
Today’s Agenda
• Hill climbing
• Simulated annealing
• Local beam search
• Genetic algorithms

Local Search Algorithms
• Local Search Algorithms keep a single "current" state,
and move to neighboring states in order to try
improve it.
• Solution path needs not be maintained.
• Hence, the search is “local”.

• Local search suitable for problems in which path is not
important; the goal state itself is the solution.
• It is an optimization search
Classical Search vs. Local Search
Classical search Local Search
• systematic exploration of • In many optimization problems,
search space. the path to the goal is irrelevant;
the goal state itself is the solution.
• Keeps one or more paths
• State space = set of "complete"
in memory.
configurations.
• Records which • Find configuration satisfying
alternatives have been constraints, Find best state
explored at each point according to some objective
along the path. function h(s). e.g., n-queens, h(s)=
number of attacking queens. In
• The path to the goal is a
such cases, we can use Local
solution to the problem.
Search Algorithms.

Example: n-queens
• Put n queens on an n × n board
with no two queens on the same
row, column, or diagonal.
• In the 8-queens problem, what
matters is the final configuration
of queens, not the order in which
they are added.

Local Search: Key Idea
Key idea:
1. Select (random) initial state (generate an initial guess).
2. Make local modification to improve current state

(evaluate current state and move to other states).
3. Repeat Step 2 until goal state found (or out of time).

Local Search: Key Idea
Advantages Drawback
• Use very little memory – • Local Search can get stuck in
usually a constant amount. local maxima and not find the
• Can often find reasonable optimal solution.
solutions in large or infinite
state spaces (e.g.,
continuous). For which
systematic search is
unsuitable.

State-Space Landscape
• A state space landscape: is a graph of states associated with their costs.
• State-space landscape
– Location (defined by state)
– Elevation (defined by the value of the heuristic cost function or objective
function)
– If elevation = cost, aim to find the lowest valley (a global minimum)
– If elevation = objective function, find the highest peak (a global maximum)
– A complete local search algorithm always find a goal if one exists
– An optimal algorithm always find a global minimum/maximum

Local and Global Optima
• Global optimum
– A solution which is better than all other
solutions
– Or no worse than any other solution
• Local optimum
– A solution which is better than nearby
solutions
– A local optimum is not necessarily a global
one
Global / Local (max/min)
• A local max/min is over a small area.

– For instance, if a point is lower than the next
nearest point on the left & right than it's a local min.
• There can be many local maxs and mins over an
entire graph.
• A global max/min is the highest/lowest point on
the entire graph.
• There can only be ONE gobal max and/or min
on a graph and there may not be one at all.
Global / Local (max/min)

Hill-Climbing Search

• Main Idea: Keep a single current node and move to a neighboring
state to improve it.
• Uses a loop that continuously moves in the direction of increasing
value (uphill):
– Choose the best successor, choose randomly if there is more
than one.
– Terminate when a peak reached where no neighbor has a
higher value.
• It also called greedy local search, steepest ascent/descent.

• “Like climbing Everest in thick fog with amnesia”
• Only record the state and its evaluation instead of
maintaining a search tree
function HILL-CLIMBING ( problem ) returns a state that is a local maximum

inputs: problem, a problem
local variables: current, a node
neighbor, a node
current  MAKE-NODE( INITIAL-STATE[ problem ])

loop do
neighbor  a highest-valued successor of current
if VALUE[ neighbor ]  VALUE[ current ] then return STATE[ current ]
current  neighbor

it is possible to make progress
attempts to find a better
solution by incrementally
changing a single element
of the solution
Plateaux
Plateaux
Amna Iftikhar
Current state Spring ' 2021 15
Local maxima: a local maximum is a peak that is
higher than each of its neighboring states, but
lower than the global maximum. Hill-climbing
algorithms that reach the vicinity of a local
maximum will be drawn upwards towards the
peak, but will then be stuck with nowhere else to
go.
Plateaux: a plateau is an area of the state space

landscape where the evaluation function is flat. It
can be a flat local maximum, from which no uphill
exit exists, or a shoulder, from which it is possible
to make progress.
Ridges: Ridges result in a sequence of local

maxima that is very difficult for greedy algorithms
to navigate. (the search direction is not towards
the top but towards the side)
Hill-Climbing Search Example
our aim is to find a path from S to M
associate
S heuristics with
every node, that
is the straight line
A B 11 distance from the
9
path terminating
city to the goal
city
7.5 C 8.5 D 8 E F 9 G 9
6 H 5 I J 7 K 6
2 L 0 M N 4 O 4

S
9 A B 11
7.5 C 8.5 D 8 E F 9 G 9
6 H 5 I J 7 K 6
2 L 0 M N 4 O 4

S
9 A B 11
7.5 C 8.5 D 8 E F 9 G 9
6 H 5 I J 7 K 6
2 L 0 M N 4 O 4

S
9 A B 11
7.5 C 8.5 D 8 E F 9 G 9
6 H 5 I J 7 K 6
2 L 0 M N 4 O 4

S
9 A B 11
7.5 C 8.5 D 8 E F 9 G 9
6 H 5 I J 7 K 6
2 L 0 M N 4 O 4

Local Maximum
From A find a solution where 10 A
H and K are final states
10 B 7 F
8 J
4 D C E 3 G
2 5
0 K
I 6
0 H
K 0

Local Maximum
10 A
10 B 7 F
8 J
4 D C E 3 G
2 5
0 K
I 6
0 H
K 0

Local Minimum
10 A
10 B 7 F
8 J
4 D C E 3 G
2 5
0 K
G is local minimum I 6
0 H
K 0
Hill climbing is sometimes called greedy local search because it grabs a good neighbor
state without thinking ahead about where
Amna Iftikhar Springto go next.
' 2021 24
Local Maximum, Local Minimum

Alternative hill climbing
• Stochastic hill climbing
– chooses at random from among the uphill moves (neighbors); the probability of
selection can vary with the steepness of the uphill move.
– This usually converges more slowly than steepest ascent, but in some state
landscapes it finds better solutions.
• First-choice hill climbing
– implements stochastic hill climbing by generating successors randomly until one is
generated that is better than the current state.
– This is a good strategy when a state has many (e.g., thousands) of successors.
• Random-restart hill climbing
– adopts the well known adage(proverb), "If at first you don't succeed, try, try again."
It conducts a series of hill-climbing searches from randomly generated initial state,
stopping when a goal is found.
– It is complete with probability approaching 1, for the trivial reason that it will
eventually generate a goal state as the initial state. (It iteratively does hill-climbing,
each time with a random initial condition)
Simulated Annealing Search

Annealing
What does the term annealing mean?
Annealing, in metallurgy and materials science, is a heat

treatment wherein a material is altered, causing changes in its
properties such as strength and hardness. It is a process that
produces conditions by heating to above the recrystallization
temperature and maintaining a suitable temperature, and then
cooling. Annealing is used to induce ductility, soften material,
relieve internal stresses, refine the structure by making it
homogeneous, and improve cold working properties.

Simulated Annealing Search
• Main Idea: escape local maxima by
allowing some "bad" moves but
gradually decrease their frequency.
• Select a neighbor at random.
• If better than current state go there.
• Otherwise, go there with some

probability.
• Probability goes down with time
(similar to temperature cooling)
Simulated annealing search
• Initialize current to starting state
• For i = 1 to 
– If T(i) = 0 return current
– Let next = random successor of current
– Let  = value(next) – value(current)
– If  > 0 then let current = next
– Else let current = next with probability exp(/T(i))

Probability of selecting bad moves
Temperature Probability of acceptance

1-worse 2-worse 3-worse
10 0.9 0.82 0.74
1 0.37 0.14 0.05
0.25 0.018 0.0003 0.000006
0.1 0.00005 2×10-9 9×10-14

Simulated annealing search
• One can prove: If temperature decreases slowly
enough, then simulated annealing search will find a
global optimum with probability approaching one
• However:
– This usually takes impractically long
– The more downhill steps you need to escape a local
optimum, the less likely you are to make all of them in a row

Local Beam Search

Local beam search
 Main Idea: Keep track of k states rather than just one.
 Start with k randomly generated states.
 At each iteration, all the successors of all k states are generated.
 If any one is a goal state, stop; else select the k best successors from the
complete list and repeat.
 Drawback: the k states tend to regroup very quickly in the same region  lack of
diversity.
 Is this the same as running k greedy searches in parallel?
Greedy search Beam search

Local beam search
• Instead of keeping only one node in memory
– Keep track of k states
– Start with k randomly generated states
– For each step, generate the successors of all k states
– If anyone is a goal, the algorithm halts
• Otherwise selects the k best successors from the complete list and
repeats

Local beam search
• Look like instead of doing it in parallel, rather than sequential, but
– In random-search, search independently of others
– In local beam search, useful info is passed among the parallel search
threads
• Quickly abandon unfruitful searches, and move to the resourceful ones
• Drawbacks
– Lack of diversity among k states
– Can quickly become concentrated in a small region of search space
(more expensive version of hill climbing)
• Alternative: Stochastic beam search
– Instead of choosing best k from the successors, choose k successors at
random (with probability of choosing a given successor being an
increasing function of its value)
– Similar to natural selection
Genetic Algorithm Search

Knapsack problem

Genetic Algorithms
• Formally introduced in the US in the 70s by John Holland.
• GAs emulate ideas from genetics and natural selection and
can search potentially large spaces.
• Before we can apply Genetic Algorithm to a problem, we
need to answer:
- How is an individual represented?

- What is the fitness function?
- How are individuals selected?
- How do individuals reproduce?

Genetic Algorithms: Representation of
states (solutions)
• Each state or individual is represented as a string over a

finite alphabet. It is also called chromosome which Contains
genes.
genes
Solution: 607 1001011111

Encoding
Chromosome:
Binary String

Genetic Algorithms: Fitness Function
• Each state is rated by the evaluation function called
fitness function. Fitness function should return
higher values for better states:
Fitness(X) should be greater than Fitness(Y) !!
[Fitness(x) = 1/Cost(x)]
Cost
States
Amna Iftikhar
X Spring ' 2021
Y 41
GA Parent Selection - Roulette Wheel
• Sum the fitnesses of all the
population members, TF
• Generate a random number, m,
between 0 and TF
• Return the first population
member whose fitness added
Roulette Wheel Selection to the preceding population
members is greater than or
equal to m

Genetic Algorithms: Selection
How are individuals selected ?
Roulette Wheel Selection
1 2 3 4 5 6 7 8
1 2 3 1 3 5 1 2
0 Rnd[0..18] = 7 Rnd[0..18] = 12 18
Chromosome4 Chromosome6
Genetic Algorithms: Cross-Over
and Mutation
How do individuals reproduce ?
Genetic Algorithms
Crossover - Recombination
1010000000 Parent1 Offspring1 1011011111

1001011111 Parent2 Offspring2 1000000000
Crossover
single point - With some high probability (crossover rate)
random apply crossover to the parents. (typical
values are 0.8 to 0.95)
Stochastic Search: Genetic Algorithms
Mutation
mutate
Offspring1 1011011111 Offspring1 1011001111
Offspring2 1010000000 Offspring2 1000000000
Original offspring Mutated offspring
With some small probability (the mutation rate) flip

each bit in the offspring (typical values between 0.1
and 0.001)
Genetic Algorithm
Example
Maxone problem
Genetic Algorithms
Algorithm:
1. Initialize population with p Individuals at
random
2. For each Individual h compute its fitness
3. While max fitness < threshold do
Create a new generation Ps
4. Return the Individual with highest fitness

Genetic algorithms
Has the effect of “jumping” to a completely different new

part of the search space (quite non-local)

Genetic Algorithm (cont.)
Initial Population Fitness Fn Selection Crossover Mutation
 Fitness function: # of non-attacking pair of queens

(min = 0, max = 8×7/2 = 28)
 Probability for selected for rep
 24/(24+23+20+11) = 31%
 23/(24+23+20+11) = 29%, etc
32752411 24748552 32748552
Genetic
Operate on state representation.
Algorithm
production of next generation random mutation

Fitness function: number of non-attacking
crossover point randomly generated
pairs of queens (min = 0, max = 8 × 7/2 = 28
 the higher the better)
24/(24+23+20+11) = 31% probability of a given pair selection
23/(24+23+20+11) = 29% etc proportional to the fitness (b)
GA is a good no clue approach to
problem solving
• GA is superb if:
– Your space is loaded with lots of weird bumps and local
minima.
• GA tends to spread out and test a larger subset of
your space than many other types of
learning/optimization algorithms.
– You don’t quite understand the underlying process of
your problem space.
– You have lots of processors
• GA’s parallelize very easily!

60
Evolvable Circuits
Amna Iftikhar Spring ' 2021

61
Robotics

62
Car Design

Trip traffic and shipment routing

Feature selection in Machine Learning

65
Evolutionary Arts
What is the major challenge?

66
Evolving Mona Lisa

Introduction To Artificial Intelligence: Amna Iftikhar Spring ' 2021 1

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Introduction To Artificial Intelligence: Amna Iftikhar Spring ' 2021 1

Uploaded by

Copyright:

Available Formats

Introduction to Artificial

Amna Iftikhar Spring ' 2021 2

• Solution path needs not be maintained.

• Hence, the search is “local”.

Amna Iftikhar Spring ' 2021 4

Amna Iftikhar Spring ' 2021 5

2. Make local modification to improve current state

Amna Iftikhar Spring ' 2021 6

Amna Iftikhar Spring ' 2021 7

Amna Iftikhar Spring ' 2021 8

• A local max/min is over a small area.

Amna Iftikhar Spring ' 2021 11

Amna Iftikhar Spring ' 2021 12

Amna Iftikhar Spring ' 2021 13

function HILL-CLIMBING ( problem ) returns a state that is a local maximum

current  MAKE-NODE( INITIAL-STATE[ problem ])

Amna Iftikhar Spring ' 2021 14

Plateaux: a plateau is an area of the state space

Ridges: Ridges result in a sequence of local

Amna Iftikhar Spring ' 2021 17

Amna Iftikhar Spring ' 2021 18

Amna Iftikhar Spring ' 2021 19

Amna Iftikhar Spring ' 2021 20

Amna Iftikhar Spring ' 2021 21

Amna Iftikhar Spring ' 2021 22

Amna Iftikhar Spring ' 2021 23

Amna Iftikhar Spring ' 2021 25

Amna Iftikhar Spring ' 2021 27

What does the term annealing mean?

Annealing, in metallurgy and materials science, is a heat

Amna Iftikhar Spring ' 2021 28

• If better than current state go there.

• Otherwise, go there with some

Amna Iftikhar Spring ' 2021 30

Temperature Probability of acceptance

Amna Iftikhar Spring ' 2021 31

Amna Iftikhar Spring ' 2021 32

Amna Iftikhar Spring ' 2021 33

 At each iteration, all the successors of all k states are generated.

Greedy search Beam search

Amna Iftikhar Spring ' 2021 35

Amna Iftikhar Spring ' 2021 37

Amna Iftikhar Spring ' 2021 38

- How is an individual represented?

Amna Iftikhar Spring ' 2021 39

• Each state or individual is represented as a string over a

Solution: 607 1001011111

Amna Iftikhar Spring ' 2021 40

Amna Iftikhar Spring ' 2021 42

Roulette Wheel Selection

1010000000 Parent1 Offspring1 1011011111

Offspring1 1011011111 Offspring1 1011001111

Offspring2 1010000000 Offspring2 1000000000

Original offspring Mutated offspring

With some small probability (the mutation rate) flip

Amna Iftikhar Spring ' 2021 55

Has the effect of “jumping” to a completely different new

Amna Iftikhar Spring ' 2021 56

Initial Population Fitness Fn Selection Crossover Mutation

 Fitness function: # of non-attacking pair of queens

production of next generation random mutation

Amna Iftikhar Spring ' 2021 59