Local Search and

Optimization Problems
Dr. Sandareka Wickramanayake

1
Outline
• Hill Climbing
• Simulated Annealing
• Local Beam Search
• Genetic Algorithms

2
Reference
• Artificial Intelligence - A Modern Approach – Chapter 4 –
Section 1

3
Search Types
• Backtracking state-space search
• Local Search and Optimization
• Constraint satisfaction search
• Adversarial search

4
Local Search and Optimization
• Previous searches: keep paths in memory and remember
alternatives so search can backtrack. The solution is a path to
a goal.
• Path may be irrelevant if only the final configuration is needed
• 8-queens
• IC design
• Network optimization
• Factory floor layout
• Job shop scheduling
• Automatic programming
• Cop planning etc.

5
Local Search
• Use a single current state and move only to neighbours.
• No track of the paths
• No track of the set of states that have been reached
• Not systematic ➔ Might never explore a portion of the search
space where the solution exists.

• Advantages:
• Use little space
• Can find reasonable solutions in large or infinite (continuous) state
spaces for which the other algorithms are unsuitable.

6
Optimization
• Local search is often suitable for optimization problems. Search
for the best state by optimizing an objective function.
• F(x) where often x is a vector of continuous or discrete values.
• Begin with a complete configuration.
• A successor of state S is S with a single element changed.
• Move from the current state to a successor state.
• Low memory requirements, because the search tree or graph is
not maintained in memory (paths are not saved).

7
Examples – 8 Queens
• Find an arrangement of 8 queens on a chess board such that
no two queens are attacking each other (A queen attacks any
piece in the same row, column, or diagonal.).
• Start with some arrangement of the queens, one per column.
• X[j]: row of queen in column j
• Successors of a state: move
one queen
• 𝐹(𝑥): # pairs attacking each
other

8
Examples – Traveling Salesman Problem
• Visit each city exactly once.
• Start with some ordering of the cities.
• State representation – order of the cities visited.
• Successor state: a change to the current ordering
• 𝐹(𝑥): length of the route

9
Comparison to Tree Search Framework
• Chapter 3: start state is typically not a complete configuration.
Chapter 4: all states are
• Chapter 3: binary goal test;
Chapter 4: no binary goal test, unless one can define one in terms of
the objective function for the problem (e.g., no attacking pairs in 8-
queens).
• Heuristic function: an estimate of the distance to the nearest goal
Objective function: preference/quality measure – how good is this state?
• Chapter 3: saving paths
Chapter 4: start with a complete configuration and make modifications
to improve it
10
Visualization
• States are laid out in a landscape.
• Elevation corresponds to the objective function value.
• Move around the landscape to find the highest (or lowest) peak.
• Only keep track of the current states and immediate neighbors.

A one-dimensional state-space landscape in

which elevation corresponds to the objective
function. The aim is to find the global
maximum.

11
Local Search Algorithms
• Strategies for choosing the state to visit next
• Hill climbing
• Simulated annealing
• Then, an extension to multiple current states:
• Genetic algorithms

12
Hill Climbing (Greedy Local Search)
• Generate nearby successor states to the current state.
• Pick the best and replace the current state with that one.
• Loop

13
Hill Climbing (Greedy Local Search)
• Generate nearby successor states to the current state.
• Pick the best and replace the current state with that one.
• Loop

https://thumbs.gfycat.com/AlienatedImmaculateFlyingsquirrel-mobile.mp4
14
Hill Climbing Search Problems

15
Hill Climbing Search Problems
• Here, we assume maximization rather than minimization.
• Local maximum: A local maximum is a peak that is higher than
each of its neighboring states but lower than the global
maximum.

Local maximum
16
Hill Climbing Search Problems
• Ridges: It is a region that is higher than its neighbors but itself
has a slope.
• It is a special kind of local maximum.
• Results in a sequence of local maxima that is very difficult for greedy
algorithms to navigate.
• Plateau: the evaluation function is flat, resulting in a random
walk.

17
Variants of Hill Climbing
• Stochastic hill climbing
• Chooses at random from among the uphill moves; 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.

18
Variants of Hill Climbing
• Random-restart hill climbing
• Start different hill-climbing searches from random starting positions
stopping when a goal is found.
• Save the best result from any search so far.
• If all states have an equal probability of being generated, it is complete
with a probability approaching 1 (a goal state will eventually be
generated).
• Finding an optimal solution becomes the question of a sufficient
number of restarts.
• Surprisingly effective, if there aren’t too many local maxima or plateau.

19
Simulated Annealing
• A hill-climbing algorithm that never makes “downhill” moves is always
vulnerable to getting stuck in a local maximum.
• A purely random walk that moves to a successor state without
concern for the value will eventually discover the global maximum
but will be extremely inefficient.
• Combine the ideas: add some randomness to hill-climbing to allow
the possibility of escape from a local optimum.

• Simulated annealing wanders around during the early parts of the

search, hopefully toward a good general region of the state space
• Toward the end, the algorithm does a more focused search, making
a few bad moves.

20
Simulated Annealing
• Based on a metallurgical metaphor.
• Annealing: harden metals and glass by heating them to a high
temperature and then gradually cooling them.
• Start with a temperature set very high and slowly reduce it.

• Run hill-climbing with the twist that you can occasionally replace
the current state with a worse state based on the current
temperature and how much worse the new state is.
• At the start, make lots of moves and then gradually slow down.

21
Simulated Annealing
• Generate a random new neighbor from the current state.
• If it’s better take it.
• If it’s worse, then take it with some probability proportional to
the temperature and the delta between the new and old states.

https://upload.wikimedia.org/wikipedia/commons/d/d5/Hill_Climbing_with_Simulated_Annealing.gif 22
 Simulated Annealing
function Simulated-Annealing(start, schedule)
current ← start
for t ← 1 to ∞ do
T ← schedule[t]
if T=0 then return current
next ← a randomly selected successor of the current
ΔE ← Value[next] – Value[current]
if ΔE > 0 then current ← next
else current ← next only with probability eΔE/T

23
Simulated Annealing
• Probability of a move decreases with the amount ΔE by which
the evaluation is worsened.
• A second parameter T is also used to determine the probability:
high T allows worse moves, T close to zero results in few or no
bad moves.
• Schedule input determines the value of T as a function of the
completed cycles.

24
Local Beam Search
• Keep track of k states rather than
just one, as in hill climbing

• Begins with k randomly generated

states
• At each step, all successors of all k
states are generated
• If anyone is a goal, algorithm halts
• Otherwise, selects the best k
successors from the complete list,
and repeats

25
Local Beam Search
• Successors can become concentrated in a small part of state
space
• Stochastic beam search: choose k successors with the
probability proportional to the successor’s value.
• Increase diversity

26
Genetic Algorithms
• Local beam search, but…
• A successor state is generated by combining two parent states.
• Start with k randomly generated states (population).
• A state is represented as a string over a finite alphabet (often a
string of 0s and 1s).
• Evaluation function (fitness function). Higher = better.
• Produce the next generation of states by selection, crossover,
and mutation.

27
Genetic Algorithms
• Representation of individuals
• Classic approach: an individual is a string over a finite alphabet, with
each element in the string called a gene
• Usually, binary instead of AGTC as in real DNA
• Mixing number
• Selection strategy
• Random
• Selection probability proportional to fitness
• Selection is made with replacement to make a very fit individual
reproduce several times

28
Genetic Algorithms
• Recombination procedure
• Random pairing of selected individuals
• Random selection of cross-over points
• Mutation rate – How often offspring have random mutations to
their representation.
• The makeup of next generation
• Forms with the newly generated offspring
• Elitism - Include a few top-scoring parents from the previous generation
in addition to newly formed offspring.
• Culling – Discard individuals below a given threshold.

29
 Example – N-Queens [16257483]

• Put n queens on an n × n board with no two

queens on the same row, column, or diagonal

fitness: #non-attacking queens Probability of being regenerated in next generation based on the 30
fitness value.
Example – N-Queens
[32752411] [24748552] [32748552]

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

the search space (quite non-local)

31
 Genetic Algorithms
function GA (pop, fitness-fn) function REPRODUCE(x,y)
Repeat n = len(x)
new-pop = {} c = random num from 1
for i from 1 to size(pop): to n
x = rand-sel(pop,fitness-fn) return APPEND
(substr(x,1,c),substr(y,
y = rand-sel(pop,fitness-fn) c+1,n)
child = REPRODUCE(x,y)
if (small rand prob): child 
mutate(child)
add child to new-pop
pop = new-pop
until an indiv is fit enough, or out of
time
return best indiv in pop, according to
fitness-fn 32
Comments on Genetic Algorithms
• Genetic Algorithm is a variant of “stochastic beam search”.
• Positive points
• Random exploration can find solutions that local search can’t (via
crossover primarily).
• Appealing connection to human evolution.
• Negative points
• Many “tunable” parameters
• Difficult to replicate performance from one problem to another
• Lack of good empirical studies comparing to simpler methods
• Useful on some (small?) problems, but no convincing evidence that
GAs are generally better than hill-climbing w/random restarts.

33