Outline. CS 6776 Evolutionary Computation. Numerical Optimization. Fitness Function. ,x 2. ) = x 2 1. , x , 5.0 x 1.

Similar documents
CS5401 FS2015 Exam 1 Key

Evolutionary Computation

Escaping Local Optima: Genetic Algorithm

CHAPTER 2 CONVENTIONAL AND NON-CONVENTIONAL TECHNIQUES TO SOLVE ORPD PROBLEM

Introduction to Optimization

Introduction to Optimization

Evolutionary Computation. Chao Lan

Evolutionary multi-objective algorithm design issues

A New Selection Operator - CSM in Genetic Algorithms for Solving the TSP

Lecture 4. Convexity Robust cost functions Optimizing non-convex functions. 3B1B Optimization Michaelmas 2017 A. Zisserman

GENETIC ALGORITHM with Hands-On exercise

DERIVATIVE-FREE OPTIMIZATION

Mutations for Permutations

Constraint Handling. Fernando Lobo. University of Algarve

Chapter 9: Genetic Algorithms

Neural Network Neurons

Evolutionary Algorithms. CS Evolutionary Algorithms 1

Multi-objective Optimization

Pseudo-code for typical EA

Optimization. Industrial AI Lab.

Genetic Algorithms for Vision and Pattern Recognition

REAL-CODED GENETIC ALGORITHMS CONSTRAINED OPTIMIZATION. Nedim TUTKUN

Hybridization EVOLUTIONARY COMPUTING. Reasons for Hybridization - 1. Naming. Reasons for Hybridization - 3. Reasons for Hybridization - 2

Artificial Intelligence Application (Genetic Algorithm)

What is GOSET? GOSET stands for Genetic Optimization System Engineering Tool

An Evolutionary Algorithm for the Multi-objective Shortest Path Problem

Reducing Graphic Conflict In Scale Reduced Maps Using A Genetic Algorithm

Optimization of Constrained Function Using Genetic Algorithm

METAHEURISTICS. Introduction. Introduction. Nature of metaheuristics. Local improvement procedure. Example: objective function

Classification of Optimization Problems and the Place of Calculus of Variations in it

Introduction to Optimization

An Introduction to Evolutionary Algorithms

CHAPTER 3.4 AND 3.5. Sara Gestrelius

Module 1 Lecture Notes 2. Optimization Problem and Model Formulation

Topic 3: GIS Models 10/2/2017. What is a Model? What is a GIS Model. Geography 38/42:477 Advanced Geomatics

The Context of Forest Management & Economics, Modeling Fundamentals. Lecture 1 (03/27/2017)

Multi-objective Genetic Algorithms for Multi-depot VRP with Time Windows

Improving interpretability in approximative fuzzy models via multi-objective evolutionary algorithms.

Evolutionary Computation for Combinatorial Optimization

Lecture on Modeling Tools for Clustering & Regression

ARTIFICIAL INTELLIGENCE (CSCU9YE ) LECTURE 5: EVOLUTIONARY ALGORITHMS

Review: Final Exam CPSC Artificial Intelligence Michael M. Richter

CS348 FS Solving NP-Complete Light Up Puzzle

NICHING EVOLUTIONARY ALGORITHMS FOR MULTIMODAL AND DYNAMIC OPTIMIZATION

Introduction to Genetic Algorithms. Genetic Algorithms

MAXIMUM LIKELIHOOD ESTIMATION USING ACCELERATED GENETIC ALGORITHMS

Genetic Algorithm Performance with Different Selection Methods in Solving Multi-Objective Network Design Problem

Suppose you have a problem You don t know how to solve it What can you do? Can you use a computer to somehow find a solution for you?

Incorporation of Scalarizing Fitness Functions into Evolutionary Multiobjective Optimization Algorithms

The k-means Algorithm and Genetic Algorithm

Multi-Objective Optimization Using Genetic Algorithms

Heuristic Optimisation

Genetic Algorithms. Kang Zheng Karl Schober

Neural Network Weight Selection Using Genetic Algorithms

Evolutionary origins of modularity

Outline of the module

The Binary Genetic Algorithm. Universidad de los Andes-CODENSA

3. Genetic local search for Earth observation satellites operations scheduling


Evolutionary Computational Approaches to Solving the Multiple Traveling Salesman Problem Using a Neighborhood Attractor Schema

Automatic Programming with Ant Colony Optimization

Multi-Objective Sorting in Light Source Design. Louis Emery and Michael Borland Argonne National Laboratory March 14 th, 2012

CHAPTER 6 REAL-VALUED GENETIC ALGORITHMS

Influence of the tape number on the optimized structural performance of locally reinforced composite structures

THE EFFECT OF SEGREGATION IN NON- REPEATED PRISONER'S DILEMMA

CS5401 FS Solving NP-Complete Light Up Puzzle

Approximate Evolution Strategy using Stochastic Ranking

METAHEURISTICS Genetic Algorithm

OPTIMIZATION METHODS. For more information visit: or send an to:

Optimizing Flow Shop Sequencing Through Simulation Optimization Using Evolutionary Methods

The Genetic Algorithm for finding the maxima of single-variable functions

CHAPTER-5 APPLICATION OF SYMBIOTIC ORGANISMS SEARCH ALGORITHM

An evolutionary annealing-simplex algorithm for global optimisation of water resource systems

Genetic Algorithms: Setting Parmeters and Incorporating Constraints OUTLINE OF TOPICS: 1. Setting GA parameters. 2. Constraint Handling (two methods)

Role of Genetic Algorithm in Routing for Large Network

Evolutionary Computation Part 2

ROBUST MULTI-OBJECTIVE OPTIMIZATION OF WATER DISTRIBUTION NETWORKS

Dr.-Ing. Johannes Will CAD-FEM GmbH/DYNARDO GmbH dynamic software & engineering GmbH

Machine Evolution. Machine Evolution. Let s look at. Machine Evolution. Machine Evolution. Machine Evolution. Machine Evolution

A Steady-State Genetic Algorithm for Traveling Salesman Problem with Pickup and Delivery

Multi-Objective Memetic Algorithm using Pattern Search Filter Methods

CONCEPT FORMATION AND DECISION TREE INDUCTION USING THE GENETIC PROGRAMMING PARADIGM

Evolutionary Algorithms: Lecture 4. Department of Cybernetics, CTU Prague.

Computational Methods. Randomness and Monte Carlo Methods

Differential Evolution

Genetic Programming Part 1

Hybrid Two-Stage Algorithm for Solving Transportation Problem

Pre-requisite Material for Course Heuristics and Approximation Algorithms

Lecture Set 1B. S.D. Sudhoff Spring 2010

Outline. Best-first search. Greedy best-first search A* search Heuristics Local search algorithms

DETERMINING MAXIMUM/MINIMUM VALUES FOR TWO- DIMENTIONAL MATHMATICLE FUNCTIONS USING RANDOM CREOSSOVER TECHNIQUES

Estimation of Distribution Algorithm Based on Mixture

A Course on Meta-Heuristic Search Methods for Combinatorial Optimization Problems

Evolutionary Algorithms and the Cardinality Constrained Portfolio Optimization Problem

Advanced Search Genetic algorithm

Heuristic Optimization Introduction and Simple Heuristics

CHAPTER 4 FEATURE SELECTION USING GENETIC ALGORITHM

Path Planning Optimization Using Genetic Algorithm A Literature Review

Clustering CS 550: Machine Learning

AN EVOLUTIONARY APPROACH TO DISTANCE VECTOR ROUTING

Transcription:

Outline CS 6776 Evolutionary Computation January 21, 2014 Problem modeling includes representation design and Fitness Function definition. Fitness function: Unconstrained optimization/modeling Constrained optimization/modeling Multi-objective optimization/modeling Relative fitness: co-evolution Population size and population models Convergence and termination criteria Fitness Function A mathematical function that quantifies how good a solution is. A problem can be modelled as a maximization or a minimization problem. Example: TSP: A tour X = {x i },i =1 n minimize f (X) = n n j =1 d ij d ij = dist(x i, x j ), 0, x i to x i is in the tour otherwise Maximize f (x 1, x 2 ) = x 2 1 + x 2 2, 5.0 x 1,x 2 5.0 How many optima? What are they? umerical Optimization 1

Regression Model Example Regression models, such as symbolic regressions, where the output is a real value, the most commonly used fitness function is minimizing error: ( y i y ˆ i ) 2 MSE: mean squared error f (x) = Biased more weighted on outliers Weights large errors more heavily than small ones y MAE: mean absolute error i y ˆ i f (x) = Less biased The true value of the outliner point y = 60: Liner model: the predicted value ŷ=80: MSE: 80-60 ^2=400 MAE: 80-60 =20 on-liner model: the predicted value ŷ=80: MSE: 80-70 ^2=100 MAE: 80-70 =10 Classification Model Classification models, where the output is a discrete label, the most commonly used fitness function is maximizing accuracy: Problematic when the data set is imbalanced. t i f (x) =, t i = 1, y i = ˆ 0, y i ˆ y i y i Constrained Optimization One way to handle solutions that violate the constraints is to assign penalty p in the fitness function: f (x)'= f (x) ± p For maximization problems, p is subtracted from f(x). For minimization problems, p is added to f(x) There are other constraints handling methods, other than sum-penalty. 2

Multi-Objective Optimization Problems require satisfying more than one objective: e.g. maximizing profit & minimizing cost. Weighted sum approach: Converts all fitness function for maximization, e.g. convert fitness by multiply -1; m λ = [w 1, w m ],w i 0, w i =1 F(x) = w i f i (x) Maintain a set of Pareto Front solutions: m Fitness is a vector F(x) = [ f 1 (x), f 2 (x), f 3 (x) f m (x)] Select Pareto-optimal solution (discussed in later lecture) Relative Fitness The fitness of an individual is measured in relation to that of other individuals in the same population or in a competing population. Interaction Patterns: Single population Relative (competitive) fitness vs. absolute fitness Multiple populations All vs. previous-best eighborhood interaction Standard Evolutionary Algorithm Individual fitness is based on its absolute performance without interacting with others. Co-evolution Single Population Individuals are evaluated by having them interact with each other, e.g. play a game. Play games with randomly selected individuals from the population 3

Co-evolution Multiple Populations In asymmetric games (the strategies for player 1 are different from that of the player 2), each member of Pop. 1 interacts with each member of Pop. 2. Evolutionary Algorithms Workflow To design an effective evolutionary algorithm, one need to consider the problem at hand. Population Size Intuitively, the population size can be viewed as a measure of the degree of parallel search an EA supports. A larger population size provides: better coverage of the search space (diversity) which helps high fitness individuals to be included in the initial population a larger past memory, so that good individuals do not lost so quickly during evolution. Population Size - Continued Depending on the complexity of the problem fitness landscape, different population size is needed to search a solution. Increase population size beyond necessary would take EA longer to find a solution. 4

Generational Model on-overlapping Population: gen-0 gen-1 gen-2 gen-n Canonical Genetic Algorithms: Parent selection only; all offspring are kept in the following generation. Evolution Strategies (µ, λ): Random parent selection to generate a large number of offspring; Select fitter offspring to form the new generation. Generational Model - Continued Under stochastic selection, an individual, regardless how fit it is, may only live for one generation, hence has a short-term impact on evolution. During stochastic selection, best solutions might get lost and are never carried over to new generation. These can be fixed by deterministic selection, or elite selection. Steady-state Model Overlapping Population pop replace parents offspring Generation Gap: The proportion of the population that is replaced [Sarma & De Jong, 1995]. Generational model: pop_size/pop_size=1.0 Steady-state model: #_replaced/pop_size Steady-state Model - Continued Offspring and parents compete to survive in the population. Fit individual can live for a long period of time to impact the evolution. A fit offspring can have impact on the evolution immediately after its birth, without waiting until the next generation. Impact evolutionary search? 5

EA Termination Criteria EA search process termination criteria: When a specified number of generation is reached. When the known best solution is found. When the population is converged : no further changes in the population may occur. Convergence Practical ways to detect convergence: Measure the degree of homogeneity of the population using spatial dispersion or entropy: When the homogeneity measure approaches 0, the population is converged. Measure the global fitness improvement: When the best fitness does not improve for a certain number of generation (typically 10-20), the population is converged. 6