Constraint Handling with EAs

Constraint Handling with EAs

CONTENTS

1. Motivation and the trouble

2. What is a constrained problem?

3. Evolutionary constraint handling

4. A selection of related work

5. Conclusions, observations, and suggestions

Motivation

Why constraints?

• Practical relevance:

a great deal of practical problems are constrained.

• Theoretical challenge:

a great deal of untractable problems (NP-hard etc.) are constrained.

Why evolutionary algorithms?• EAs show a good ratio of (implementation) effort/performance.

• EAs are acknowledged as good solvers for tough problems.

The trouble with constraints and EAs

• Mutating a feasible individual can result in an infeasible new individual.

• Recombining two feasible individuals can result in infeasible new individuals.

Standard reproduction operators are blind to constraints.

What is a constrained problem?

Consider the Travelling Salesman Problem for n cities:

• Let C = {city1, … , cityn}

If we define the search space as

• S = Cn, then we need a constraint requiring uniqueness of

each city in an element of S

• S = {permutations of C}, then we need no additional

constraint.

The notion ‘constrained problem’ depends on what we take as

search space

What is constrained / free search?

• Even in the space S = {permutations of C} we cannot perform free search

• Free search: standard mutation and crossover preserve membership of S : mut() S and cross(, t) S

• The notion ‘free search’ depends on what we take as standard mutation and crossover.

• Standard mutation of x1, … , xn:mut(x1, … , xn) = x’1, … , x’n, x’i domain(i)

• Standard crossover of x1, … , xn, y1, … , yn (and ...) if for each child z1, … , zn:• zi {xi, yi} discrete case • zi [xi, yi] continuous case

Optimization Problems (FOPs) Free search space: S = D1 … Dn

• one assumption on Di: if it is continuous, it is convex

• the restriction si Di is not a constraint, it is the definition of the domain of the i-th variable

• membership of S is coordinate-wise, hence

• a free search space allows free search

Free Optimization Problem: S, f, •• S is a free search space

• f is a (real valued) objective function on S

• Solution of an FOP: S such that f () is optimal in S

FOPs are ‘easy’, in the sense that:• it's ‘only’ optimizing, no constraints and

• EAs have a basic instinct for optimization

Constrained Optimization Problems (COPs) Constrained Optimization Problem: S, f, Φ• S is a free search space • f is a (real valued) objective function on S • Φ is a formula (Boolean function on S)

Solution of a COP: SΦ such that f() is optimal in SΦ

• SΦ = { S | Φ() = true} is the feasible search space

Φ is the feasibility condition – may be a combination of several constraints

• In general with m constraints they will fall into two types: • inequalities: g_i(x) < 0 1…i…q• equalities h_i(x) = 0 q+1 … i…m

Feasible & Unfeasible Regions

• In general S will be split into two disjoint sets of spaces:

– F (the feasible regions)

– U (the unfeasible regions).

• The spaces in F may be connected or not.

S

X

U

F

ns

Constraint Satisfaction Problems (CSPs)

Constraint Satisfaction Problem: S, •, Φ• S is a free search space • Φ is a formula (Boolean function on S) Solution of a CSP: S such that Φ() = true

Φ is typically given by a set (conjunction) of constraints

ci = ci(xj1, … , xjni), where ni is the arity of ci

ci Dj1 … Djni is also a common notation

FACTS: • the general CSP is NP-complete • every CSP is equivalent with a binary CSP (ni 2)Note that there is nothing to be optimized in a CSP

Examples

• FOP: Schwefel function

– S = D1 … Dn Di = [-512.03, 511.97]

– f() = 418.9829n -

• COP: travelling salesman problem– S = Cn, C = {city1, … , cityn}

– Φ() = true i, j {1, … , n} i j si sj

– f() =

• CSP: graph 3-coloring problem– G = (N, E), E N N , |N| = n

– S = Dn, D = {1, 2, 3}

– Φ() = ΛeE ce(), where ce() = true iff e = (k, l) and sk sl

n

iii xx

1

||sin

n

inii ssssdist

1111 : where),,(

Solving CSPs by EAs (1)

• EAs can only work if they have an f to optimize.

Thus a CSP S, •, Φ must be transformed first to a

Case 1. FOP: S, •, Φ S, f, •Case 2. COP: S, •, Φ S, f, Φ

• The transformation must be (semi-)equivalent,

i.e. at least the following implications must hold:

1. f () is optimal in S Φ() 2. ψ () and f () is optimal in S Φ()

• Case 1: All constraints are handled indirectly, i.e. they are solved by `simply‘ optimizing in S, f, •

e.g. f is number of violated constraints (to be minimized)

• Case 2: Some constraints are handled indirectly (those transformed into f), some constraints are handled directly (those that remain constraints in ψ)

Solving CSPs by EAs (2)

Constraint handling has two meanings:

• 1. how to transform the constraints in Φ into f, respectively f, ψ before applying an EA

• 2. how to enforce the constraints in S, f, Φ while running an EA

Case 1: constraint handling only in the 1st sense (pure penalty approach)

Case 2: constraint handling in both senses

In Case 2 the question

‘How to solve CSPs by EAs’

transforms to

‘How to solve COPs by EAs’

Direct constraint handling

Notes:

• only needed in Case 2, for some constraints

• feasible here means Ψ in S, f, Ψ, not Φ in S, •, Φ

‘Standard’ list (minus penalty functions):

• eliminating infeasible candidates (very inefficient, hardly practicable)

• repairing infeasible candidates

• preserving feasibility by special operators(requires feasible initial population, which can be NP-hard,

e.g. for TSP with time windows)

• decoding, i.e. transforming the search space(allows usual representation and operators)

Indirect vs. Direct Handling• Indirect PRO’s

• conceptually simple, transparent

• problem independent

• reduces problem to ‘simple’ optimization

• allows user to tune on his/her preferences by weights

• allows EA to tune fitness function by modifying weights during the search

CON’s:

• loss of info by packing everything in a single number

• said not to work well for sparse problems

• DirectPRO's:

• it works well

• (except eliminating)

CON's

• problem specific

• no guidelines

Another categorization of constraint handling

In it ia lisa tio n a n d va ria tiono p e ra to rs

d e s ign e d so the y a lw a ysg ive va lid so lu tio ns

G e no typ e ->p he n otypeM a pp ing

e n su re s fe as ib ilityv ia de cod er

O n ly Fe a s ib le S o lu tio nsA llo w ed

S o lu tio ns pe n a liseda cco rd in g to th e e xte n t o f

co n stra in tv io la tion

"B a ld w in ia n"in fe as ib le so lu tionke p t w ith fitn e sso f fea s ib le po in t

"L a m ark ia n""re p a ire d"

fe a sib le so lu tio n ke p tin po pu la tion

"R e p a ir" fu nc tio nsu sed to g en era tefe a sib le so lu tio ns

c lo se to in fea s ib le po in t

In fea s ib le S o lu tio nsA llo w ed

C o n stra in ed O p tim isa tion

PRO-LIFEPRO-CHOICE

Penalty Functions

• Assuming a minimisation problem, the general form, will be :

min f(x) such that x A, x Bwhere the constraints x A are easy to satisfy and

those x B are harder

• This can be reformulated as:min f (x + p( d( x, B) ) where

d( x, B) is a distance function of x from B , andp(.) is a penalty function s.t. p(0) =0

Issues with penalty functions

• Distance metrics might include: • number of violated constraints• sum of constraint violations ( linear or

exponential)• distance to feasible region

• Penalty function must balance exploration of infeasible region, with not wasting time

Static Penalty Functions

Some variants:

1. Fixed penalty for all infeasible solutions

2. Add a penalty C for each violated constraint– These two are generally inferior to using some distance

metric from feasible region

3. Penalty Functions based on cost of repair:

wherei

m

iip dCxfxf

1

)()(

)(

)(

xh

xgd

i

iii

Dynamic Penalty Functions

• Attempt to solve problem of setting the values of C_i by hand.

• Idea is to progressively penalise infeasible solutions as time progresses

• Still need to fine tune penalties to problem– too lenient => infeasible solutions in final pop.– too harsh => converge to sub-optimal solution

Adaptive Penalty Functions

• Bean & Hadj-Alouane: – revise penalties based on performance of best

over last few generations.

• Smith & Tate / Coit et al.– penalty involves estimating “near feasible

threshold” over time

• Eiben et al. – change bias as constraints satisfied

Decoders• Methods of dealing with CSP’s where

genome gives instructions to decoder on how to build feasible solution.

• Decoder maps from a Representation space R into the feasible regions F of the solution space S

• for every solution s in F there must be at least one representation d in R

• every d in R must map to one solution s in S

• every s in S must map to the same number of representations in R : need not be 1:1

Decoder Example: Knapsack Problem• Sort items by profit/weight• Decoder reads string from left to right and

interprets a ‘1’ as “take this item if possible” e.g. 11011000100100…..

is interpreted as take items 1,2,5,6, 10 …

• Redundancy arises at right hand end of string.• Can use normal crossover and mutation operators

with guarantee of feasible solutions

Decoder Example: TSP• Representation: alleles in position i can have

values in the range [1, n - i +1] e.g. 1:1-9, 8:1,2• Decoder views Cities as ordered list, starting at

position 1 with C = (1,2,3,4,5,6,7,8,9)

• In position i put the d_ith element of C and then remove that element from Ce.g. 124386967 is represented by [112141311]

• Can use normal crossover and (relatively) normal mutation operators

Repair Algorithms• Method of solving CSP’s where an algorithm is

used to “repair” infeasible solutions, i.e:eval_u (x) = eval_f(y)

where y is repaired version of x

• Major issue of whether to replace x with y– Baldwin vs. Lamarckian learning.

– Some authors suggest P(replacement) ~ 5-15%

• For some problems repair algorithm may be very time consuming

Example Repair Algorithms:Discrete

• For instance in Knapsack problem with overfull knapsack, algorithm selects and removes items until weight constraint is met.

• Choice of item to remove could be:– random probabilistic

– leftmost/rightmost deterministic– one with worst p/w ratio deterministic “greedy”– using p.d.f. according to p/w ratio probabilistic

Example Repair: Continuous

• Genocop Algorithm (Michalewicz)– Maintains a population of solutions Ps and a population

of feasible “reference points” Pr

– Evaluation of infeasible solutions depends on contents of Pr which adapt over time.

– infeasible s evaluated by picking r from Pr

• Draw segment from s to r

• select sequence of points along segment

• Stop at first feasible point z : f(s) = f(z)

• If f(z) > f(r) replace r in Pr with z

Genocop Options

• Genocop has three parameters:

– selection of points along segment rs

z = a_i(r) + (1-a_i)(s)

a_i may be randomly chosen or (0.5, 0.25, 0.125…)

– choice of reference point r may be fittest in Pr or

random

– probability of replacing infeasible solution s with

repaired version z (0.15 seems good …)

Multi-Objective Approaches to CSP

• Each of the m constraints has a violation measure f_m(x) associated with it

• Treat the fitness function and these m functions as a m+1 dimensional vector whose components need to be minimised

• Use multi-objective evolutionary approaches such as VEGA (Schaffer 1984)

• successfully applied to problems such as Gas-pipeline layout

Co-evolutionary Approaches to CSP

• Paredis (1994)• population of solutions + population of constraints• fitter solutions violate less constraints• fitter constraints are violated by more solutions

• Segregated GA (Le Riche et al. 1995)• two populations with different penalty functions• rank & merge =>1 pop => generate offspring• evaluate offspring with each fcn =>2 pops• repeat• Balances penalty functions => more robust search

Summary of some research results

Global picture arising

Two principal approaches:

1 heuristics for mutation, crossover, repair, fitness function2 time varying fitness (penalty) functions

Straightforward research questions:

• which option is better (on a given problem class)• what if the two options are combined

Some general observations/conclusions

• CSP FOP is common

• CSP COP is not common

• Mutation only and small populations work well on CSPs

• Presence of constraints facilitates natural measures on sub-individual structures (kind of fitness values evaluating a part of a candidate solution)

• Measures on sub-individual structures allow

– CSP specific, but problem independent heuristics

– composed fitness functions

• Composed fitness functions can be naturally varied during a run explicitly or implicitly (co-evolution)

Some general observations/conclusions

Summarizing

• Incorporation of heuristics can highly increase EA performance

• Varying the fitness function can highly increase EA performance

• EAs can outperform the best heuristic benchmark techniques

CSPs can be solved by EAs