ALPS Configuration

Start here for ALPS GA

This are additional parameters that will be needed for setting up ALPS

GENERAL parameters [Use number of generations or number of evaluations]

• generations = 1000
• evaluations = 1000000
• #population size per layer
• population-size = 50

ALPS configuration

• alps-age-gap = 10
• alps-number-of-layers = 5
• alps-selection-pressure = 0.8
• alps-ss-selection-pressure = 1.0
• alps-aging-scheme = ec.algorithms.alps.agingscheme.Polynomial
• #or
• alps-aging-scheme = ec.algorithms.alps.agingscheme.exponential
• #or
• alps-aging-scheme = ec.algorithms.alps.agingscheme.Linear
• #or
• alps-aging-scheme = ec.algorithms.alps.agingscheme.Fibonacci

LAYER Replacement Strategy

• alps-replacement-strategy = ec.algorithms.alps.replacement.age.Worst
• #or
• alps-replacement-strategy = ec.algorithms.alps.replacement.age.Nearest
• #or
• alps-replacement-strategy = ec.algorithms.alps.replacement.age.ReverseTournamentWorst
• #or
• alps-replacement-strategy = ec.algorithms.alps.replacement.age.ReverseTournamentNearest prefered
• #or
• alps-replacement-strategy = ec.algorithms.alps.replacement.age.Random

The chromosomes are fixed length structures and doesnt change to mutation and crossover. The initialiser will random create integer strings in this case of length specified initial-chromosome-size = 280

EC Main Class

• main-class = ec.algorithms.alps.system.Engine

REPLACEMENT Strategy

• replacement-operation = ec.operator.operations.replacement.Generational
• #or
• replacement-operation = ec.operator.operations.replacement.SteadyState

When using Steady State Replacement, additional paraemters such as these will be needed

• steady-state-replacement = ec.operator.operations.replacement.steadystate.ReverseTournamentWorst prefered
• #or
• steady-state-replacement = ec.operator.operations.replacement.steadystate.Worst
• #or
• steady-state-replacement = ec.operator.operations.replacement.steadystate.Nearest
• #or
• steady-state-replacement = ec.operator.operations.replacement.steadystate.ReverseTournamentNearest
• #or
• steady-state-replacement = ec.operator.operations.replacement.steadystate.Random