Evolutionary Robotics

Size: px
Start display at page:

Download "Evolutionary Robotics"

Transcription

1 Evolutionary Robotics

2 Previously on evolutionary robotics

3 Evolving Neural Networks How do we evolve a neural network?

4 Evolving Neural Networks How do we evolve a neural network? One option: evolve the weights as a vector of numbers

5 Evolving Neural Networks How do we evolve a neural network? One option: evolve the weights as a vector of numbers Crossover is easy

6 Evolving Neural Networks How do we evolve a neural network? One option: evolve the weights as a vector of numbers Crossover is easy But is it sensible?

7 Evolving Neural Networks Competing conventions problem How do you cross these?

8 Evolving Neural Networks How do we evolve a neural network? One option: evolve the weights as a vector of numbers Do we have right topology for the task? To few nodes: cannot solve the task To many nodes: search space becomes unnecessarily large overfitting

9 Evolving Neural Networks How do we evolve a neural network? Another option: evolve both weights and topology But where do we start? What about crossover?

10 NEAT Neuro-Evolution of Augmenting Topologies (NEAT) Stanley & Miikkulainen Complexification 2. Intelligent crossover (via historical markings ) 3. Diversity ( speciation )

11 NEAT: Complexification Simplicity usually good Occam s razor generalizes better easier to search for computationally faster In neural nets: simple = fewer nodes and/or connections

12 NEAT: Complexification So, start simple; add complexity if needed Gen 0 = no hidden nodes Tend to (slowly) mutate in nodes across generations

13 NEAT: Intelligent Crossover Provides one solution to competing conventions problem Uses historical markings

14 NEAT: Encoding Direct Encoding

15 NEAT: Mutation add connection between any unconnected nodes - observing constraints, e.g. feedfoward add node in existing connection minimize initial effect - weight to new node is 1 - weight from new nodes is old weight

16 NEAT: Historical Innovation Numbers added for each new connection global counter for population

17 NEAT: Intelligent Crossover gene possibilities matching - inherited randomly disjoint (non-matching, middle) - inherited from more-fit parent excess (non-matching, end) - inherited from more-fit parent

18 NEAT: Diversity Newly added structure usually does not survive smaller structures are optimized faster new structure usually initially deleterious: it needs time to be tuned before having to compete Solution: protect it More generally, having diverse neural structures is a good idea

19 NEAT: Diversity NEAT uses speciation, but how do we determine which individuals belong to different species. We need a distance metric!

20 Genetic distance problem Genetic distance problem with same topology, distance is easy distance? but with different topologies? distance?

21 and now, we continue.

22 NEAT: Genetic Distance uses historical markings more disjoint & excess genes = less shared evolutionary history also include weight differences Genetic Distance Disjoint Genes Avg. Weight Difference in Matched Genes Normalizer: numgenes in larger genome C1, C2, & C3 weight the relative importance of each component

23 NEAT: Speciation Uses distance threshold If new org is <threshold away from any current species, add it else: create new species

24 NEAT: Speciation Each species is represented by a random genome inside the species in a previous generation

25 NEAT: Speciation Each species is represented by a random genome inside the species in a previous generation

26 NEAT: Speciation An individual is added to the first species within threshold 1 2 3

27 NEAT: Fitness Sharing & Selection adjusted fitness = fitness/numorgsinspecies species get numoffspring allocated by sum(adjustedfitnessofmembers) within a species, rank selection determines parents parents then mutated and crossed - crossover within a species only

28 NEAT: Summary 1. Complexification 2. Intelligent crossover (via historical markings ) 3. Diversity ( speciation ) Each component shown to help (Stanley & Miikkulainen 2005)

29 NEAT Would this make a good/fun homework exercise? Would probably not involve speciation. Historical markings? Crossover?

30 Encodings (for real this time)

31 Evolutionary Algorithms (EAs) Encode Problem genome Generate Population mutation and/or recombination W L Select Parents Score Population

32 Encodings how information is stored in a genome + process that produces phenotype gattaca ccatgat tggacct

33 Direct vs. Generative Encodings Direct Encoding: each genotypic element specifies an independent phenotypic element Genotype Phenotype leg 1: 2 leg 2: 2 leg 3: 2 leg 4: 2

34 Direct vs. Generative Encodings Direct Encoding: each genotypic element specifies an independent phenotypic element Genotype leg 1: 2 leg 2: 2 leg 3: 2 leg 4: 2 Phenotype Genotype' Phenotype' leg 1: 2 leg 2: 2 leg 3: 1 leg 4:.5

35 Direct vs. Generative Encodings Direct Encoding: each genotypic element specifies an independent phenotypic element Genotype Phenotype Genotype' Phenotype' leg 1: 2 leg 2: 2 leg 3: 2 leg 4: 2 leg 1: 2 leg 2: 2 leg 3: 1 leg 4:.5 X

36 Direct vs. Generative Encodings Direct Encoding: each genotypic element specifies an independent phenotypic element Genotype Phenotype Genotype' Phenotype' leg 1: 2 leg 2: 2 leg 3: 2 leg 4: 2 leg 1: 2 leg 2: 2 leg 3: 1 leg 4:.5 X Generative Encoding: genotypic elements can influence many phenotypic elements Genotype Phenotype 4x leg: 2

37 Direct vs. Generative Encodings Direct Encoding: each genotypic element specifies an independent phenotypic element Genotype Phenotype Genotype' Phenotype' leg 1: 2 leg 2: 2 leg 3: 2 leg 4: 2 leg 1: 2 leg 2: 2 leg 3: 1 leg 4:.5 X Generative Encoding: genotypic elements can influence many phenotypic elements Genotype Phenotype Genotype' Phenotype' 4x leg: 2 4x leg: 1

38 Nature s Encoding

39 Generative Encodings Desirable properties Coordinated mutational effects Scalability Low dimensional search, highly complex phenotype Structural Organization Regularity...with and without variation Modularity Hierarchy

40 Regularity reuse of information irregularity irregular compressibility less compressible less compressible Lipson (2007)

41 Regularity multiple regularities

42 Examples of Regularity in Generative Encodings Direct Encoding Generative Encoding Hornby (2005)

43 Examples of Regularity in Generative Encodings Direct Encoding (probably) Generative Encoding (probably)

44 Previous Work Generative outperforms direct on regular problems No tests across a continuum of problem regularity

45 Game Plan Case-study: generative encoding vs. direct encoding as problem regularity varies HyperNEAT Has a good direct encoding control Based on an important concept from developmental biology 2011

46 How nature builds complexity Generative encoding......where cell fate is a function of geometric position

47 How nature builds complexity Development involves producing complex coordinate frames Sean Carroll: Endless Forms Most Beautiful (2005)

48 How nature builds complexity

49 Compositional Pattern Producing Networks (CPPNs) Stanley 2007 encodes phenotypic elements as a function of their geometric location

50 Compositional Pattern Producing Networks (CPPNs) Stanley 2007 x y genome f(x,y) = fate y x value at x,y... for all x,y coordinates Adapted from: Stanley (2007)

51 sine(y) Compositional Pattern Producing Networks (CPPNs) Stanley 2007 x y gaussian(x) y x asymmetry f(x) left-right f(y) anterior-posterior symmetry gaussian(x) proximal-distal repetition within segment symmetric anterior-posterior value at x,y... for all x,y coordinates Adapted from: Stanley (2007)

52 Compositional Pattern Producing Networks (CPPNs) Stanley 2007 picbreeder.org

53 Previous Generative Encodings Sims 1994 Hornby & Pollack 2002 Dawkins 1986

NEUROEVOLUTION. Contents. Evolutionary Computation. Neuroevolution. Types of neuro-evolution algorithms

NEUROEVOLUTION. Contents. Evolutionary Computation. Neuroevolution. Types of neuro-evolution algorithms Contents Evolutionary Computation overview NEUROEVOLUTION Presenter: Vlad Chiriacescu Neuroevolution overview Issues in standard Evolutionary Computation NEAT method Complexification in competitive coevolution

More information

Competitive Coevolution through Evolutionary Complexification

Competitive Coevolution through Evolutionary Complexification Competitive Coevolution through Evolutionary Complexification Kenneth O. Stanley and Risto Miikkulainen Department of Computer Sciences The University of Texas at Austin Austin, TX 78712 USA kstanley,

More information

Forecasting & Futurism

Forecasting & Futurism Article from: Forecasting & Futurism December 2013 Issue 8 A NEAT Approach to Neural Network Structure By Jeff Heaton Jeff Heaton Neural networks are a mainstay of artificial intelligence. These machine-learning

More information

The Effect of Connection Cost on Modularity in Evolved Neural Networks

The Effect of Connection Cost on Modularity in Evolved Neural Networks The Effect of Connection Cost on Modularity in Evolved Neural Networks Jessica Lowell 1 and Jordan Pollack 1 1 DEMO Lab, School of Computer Science, Brandeis University, Waltham, MA 02453 jessiehl@cs.brandeis.edu

More information

Continual Coevolution through Complexification

Continual Coevolution through Complexification Continual Coevolution through Complexification Kenneth O. Stanley Department of Computer Sciences University of Texas at Austin Austin, TX 78712 kstanley@cs.utexas.edu Risto Miikkulainen Department of

More information

Neuroevolution for sound event detection in real life audio: A pilot study

Neuroevolution for sound event detection in real life audio: A pilot study Neuroevolution for sound event detection in real life audio: A pilot study Christian Kroos & Mark D. Plumbley Centre for Vision, Speech and Signal Processing (CVSSP), University of Surrey, UK Background

More information

Exploiting Regularity Without Development

Exploiting Regularity Without Development Exploiting Regularity Without Development Kenneth O. Stanley School of Electrical Engineering and Computer Science The University of Central Florida, Orlando, FL 32816 USA kstanley@cs.ucf.edu To appear

More information

Artificial Metamorphosis: Evolutionary Design of Transforming, Soft-Bodied Robots

Artificial Metamorphosis: Evolutionary Design of Transforming, Soft-Bodied Robots Copyright of figures and other materials in the paper belongs original authors. Artificial Metamorphosis: Evolutionary Design of Transforming, Soft-Bodied Robots Michał Joachimczak et al. Artificial Life

More information

Evolving Neural Networks through Augmenting Topologies

Evolving Neural Networks through Augmenting Topologies Evolving Neural Networks through Augmenting Topologies Kenneth O. Stanley kstanley@cs.utexas.edu Department of Computer Sciences, The University of Texas at Austin, Austin, TX 78712, USA Risto Miikkulainen

More information

Embodied Embeddings for Hyperneat

Embodied Embeddings for Hyperneat Embodied Embeddings for Hyperneat Collin Cappelle, Josh Bongard The University of Vermont, Burlington, VT 05405 collin.cappelle@uvm.edu Abstract A long time goal of evolutionary roboticists is to create

More information

Co-evolution of Morphology and Control for Roombots

Co-evolution of Morphology and Control for Roombots Co-evolution of Morphology and Control for Roombots Master Thesis Presentation Ebru Aydın Advisors: Prof. Auke Jan Ijspeert Rico Möckel Jesse van den Kieboom Soha Pouya Alexander Spröwitz Co-evolution

More information

Coevolution of Multiagent Systems using NEAT

Coevolution of Multiagent Systems using NEAT Coevolution of Multiagent Systems using NEAT Matt Baer Abstract This experiment attempts to use NeuroEvolution of Augmenting Topologies (NEAT) create a Multiagent System that coevolves cooperative learning

More information

Chapter 8: Introduction to Evolutionary Computation

Chapter 8: Introduction to Evolutionary Computation Computational Intelligence: Second Edition Contents Some Theories about Evolution Evolution is an optimization process: the aim is to improve the ability of an organism to survive in dynamically changing

More information

Using Genetic Algorithms to Evolve Artificial Neural Networks

Using Genetic Algorithms to Evolve Artificial Neural Networks Colby College Digital Commons @ Colby Honors Theses Student Research 2016 Using Genetic Algorithms to Evolve Artificial Neural Networks William T. Kearney Colby College Follow this and additional works

More information

V. Evolutionary Computing. Read Flake, ch. 20. Genetic Algorithms. Part 5A: Genetic Algorithms 4/10/17. A. Genetic Algorithms

V. Evolutionary Computing. Read Flake, ch. 20. Genetic Algorithms. Part 5A: Genetic Algorithms 4/10/17. A. Genetic Algorithms V. Evolutionary Computing A. Genetic Algorithms 4/10/17 1 Read Flake, ch. 20 4/10/17 2 Genetic Algorithms Developed by John Holland in 60s Did not become popular until late 80s A simplified model of genetics

More information

biologically-inspired computing lecture 18

biologically-inspired computing lecture 18 Informatics -inspired lecture 18 Sections I485/H400 course outlook Assignments: 35% Students will complete 4/5 assignments based on algorithms presented in class Lab meets in I1 (West) 109 on Lab Wednesdays

More information

Acquiring Evolvability through Adaptive Representations

Acquiring Evolvability through Adaptive Representations Acquiring Evolvability through Adaptive Representations Joseph Reisinger and Risto Miikkulainen Department of Computer Sciences The University of Texas at Austin Austin, TX 787 USA {joeraii,risto}@cs.utexas.edu

More information

V. Evolutionary Computing. Read Flake, ch. 20. Assumptions. Genetic Algorithms. Fitness-Biased Selection. Outline of Simplified GA

V. Evolutionary Computing. Read Flake, ch. 20. Assumptions. Genetic Algorithms. Fitness-Biased Selection. Outline of Simplified GA Part 5A: Genetic Algorithms V. Evolutionary Computing A. Genetic Algorithms Read Flake, ch. 20 1 2 Genetic Algorithms Developed by John Holland in 60s Did not become popular until late 80s A simplified

More information

Generative Encoding for Multiagent Learning

Generative Encoding for Multiagent Learning Generative Encoding for Multiagent Learning In: Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2008). New York, NY: ACM, 2008 (8 pages) Winner of the Best Paper Award in the

More information

Evolutionary Design I

Evolutionary Design I Evolutionary Design I Jason Noble jasonn@comp.leeds.ac.uk Biosystems group, School of Computing Evolutionary Design I p.1/29 This lecture Harnessing evolution in a computer program How to construct a genetic

More information

IV. Evolutionary Computing. Read Flake, ch. 20. Assumptions. Genetic Algorithms. Fitness-Biased Selection. Outline of Simplified GA

IV. Evolutionary Computing. Read Flake, ch. 20. Assumptions. Genetic Algorithms. Fitness-Biased Selection. Outline of Simplified GA IV. Evolutionary Computing A. Genetic Algorithms Read Flake, ch. 20 2014/2/26 1 2014/2/26 2 Genetic Algorithms Developed by John Holland in 60s Did not become popular until late 80s A simplified model

More information

Evolutionary Plantographics

Evolutionary Plantographics Evolutionary Plantographics Abstract This letter describes an evolutionary system for creating lifelike three-dimensional plants and flowers, our main goal being the facilitation of producing realistic

More information

COMP3411: Artificial Intelligence 7a. Evolutionary Computation

COMP3411: Artificial Intelligence 7a. Evolutionary Computation COMP3411 14s1 Evolutionary Computation 1 COMP3411: Artificial Intelligence 7a. Evolutionary Computation Outline Darwinian Evolution Evolutionary Computation paradigms Simulated Hockey Evolutionary Robotics

More information

SimAnt Simulation Using NEAT

SimAnt Simulation Using NEAT SimAnt Simulation Using NEAT Erin Gluck and Evelyn Wightman December 15, 2015 Abstract In nature, ants in a colony often work together to complete a common task. We want to see if we can simulate this

More information

Evolutionary Computation

Evolutionary Computation Evolutionary Computation - Computational procedures patterned after biological evolution. - Search procedure that probabilistically applies search operators to set of points in the search space. - Lamarck

More information

Evolvability, Complexity and Scalability of Cellular Evolutionary and Developmental Systems

Evolvability, Complexity and Scalability of Cellular Evolutionary and Developmental Systems 1 Evolvability, Complexity and Scalability of Cellular Evolutionary and Developmental Systems Stefano Nichele February 18, 2015 2 Outline Motivation / Introduction Research Questions Background Results

More information

Bounded Approximation Algorithms

Bounded Approximation Algorithms Bounded Approximation Algorithms Sometimes we can handle NP problems with polynomial time algorithms which are guaranteed to return a solution within some specific bound of the optimal solution within

More information

Does Aligning Phenotypic and Genotypic Modularity Improve the Evolution of Neural Networks?

Does Aligning Phenotypic and Genotypic Modularity Improve the Evolution of Neural Networks? Does Aligning Phenotypic and Genotypic Modularity Improve the Evolution of Neural Networks? Joost Huizinga, Jean-Baptiste Mouret, Jeff Clune To cite this version: Joost Huizinga, Jean-Baptiste Mouret,

More information

Genetic Engineering and Creative Design

Genetic Engineering and Creative Design Genetic Engineering and Creative Design Background genes, genotype, phenotype, fitness Connecting genes to performance in fitness Emergent gene clusters evolved genes MIT Class 4.208 Spring 2002 Evolution

More information

Genetic Algorithms. Donald Richards Penn State University

Genetic Algorithms. Donald Richards Penn State University Genetic Algorithms Donald Richards Penn State University Easy problem: Find the point which maximizes f(x, y) = [16 x(1 x)y(1 y)] 2, x, y [0,1] z (16*x*y*(1-x)*(1-y))**2 0.829 0.663 0.497 0.331 0.166 1

More information

Adaptive Generalized Crowding for Genetic Algorithms

Adaptive Generalized Crowding for Genetic Algorithms Carnegie Mellon University From the SelectedWorks of Ole J Mengshoel Fall 24 Adaptive Generalized Crowding for Genetic Algorithms Ole J Mengshoel, Carnegie Mellon University Severinio Galan Antonio de

More information

Evolutionary computation

Evolutionary computation Evolutionary computation Andrea Roli andrea.roli@unibo.it DEIS Alma Mater Studiorum Università di Bologna Evolutionary computation p. 1 Evolutionary Computation Evolutionary computation p. 2 Evolutionary

More information

Non-Adaptive Evolvability. Jeff Clune

Non-Adaptive Evolvability. Jeff Clune Non-Adaptive Evolvability! Jeff Clune Assistant Professor Computer Science Evolving Artificial Intelligence Laboratory Evolution Fails to Optimize Mutation Rates (though it would improve evolvability)

More information

Evolutionary Computation: introduction

Evolutionary Computation: introduction Evolutionary Computation: introduction Dirk Thierens Universiteit Utrecht The Netherlands Dirk Thierens (Universiteit Utrecht) EC Introduction 1 / 42 What? Evolutionary Computation Evolutionary Computation

More information

Complexity-based speciation and genotype representation for neuroevolution

Complexity-based speciation and genotype representation for neuroevolution Complexity-based speciation and genotype representation for neuroevolution Alexander Hadjiivanov School of Computer Science and Engineering University of New South Wales Sydney, NSW 0, Australia Email:

More information

Scaling Up. So far, we have considered methods that systematically explore the full search space, possibly using principled pruning (A* etc.).

Scaling Up. So far, we have considered methods that systematically explore the full search space, possibly using principled pruning (A* etc.). Local Search Scaling Up So far, we have considered methods that systematically explore the full search space, possibly using principled pruning (A* etc.). The current best such algorithms (RBFS / SMA*)

More information

odneat: An Algorithm for Decentralised Online Evolution of Robotic Controllers

odneat: An Algorithm for Decentralised Online Evolution of Robotic Controllers odneat: An Algorithm for Decentralised Online Evolution of Robotic Controllers Fernando Silva fsilva@di.fc.ul.pt Bio-inspired Computation and Intelligent Machines Lab, 1649-026 Lisboa, Portugal Instituto

More information

Towards Synthesizing Artificial Neural Networks that Exhibit Cooperative Intelligent Behavior: Some Open Issues in Artificial Life Michael G.

Towards Synthesizing Artificial Neural Networks that Exhibit Cooperative Intelligent Behavior: Some Open Issues in Artificial Life Michael G. Towards Synthesizing Artificial Neural Networks that Exhibit Cooperative Intelligent Behavior: Some Open Issues in Artificial Life Michael G. Dyer Computer Science Department, UCLA Overview Introduction

More information

COMP3411: Artificial Intelligence 10a. Evolutionary Computation

COMP3411: Artificial Intelligence 10a. Evolutionary Computation COMP3411 16s1 Evolutionary Computation 1 COMP3411: Artificial Intelligence 10a. Evolutionary Computation Outline Darwinian Evolution Evolutionary Computation paradigms Simulated Hockey Evolutionary Robotics

More information

Problems for 3505 (2011)

Problems for 3505 (2011) Problems for 505 (2011) 1. In the simplex of genotype distributions x + y + z = 1, for two alleles, the Hardy- Weinberg distributions x = p 2, y = 2pq, z = q 2 (p + q = 1) are characterized by y 2 = 4xz.

More information

Evolving Novel Cellular Automaton Seeds Using Computational Pattern Producing Networks (CPPN) *

Evolving Novel Cellular Automaton Seeds Using Computational Pattern Producing Networks (CPPN) * Evolving Novel Cellular Automaton Seeds Using Computational Pattern Producing Networks (CPPN) * Josh Wolper 1, George Abraham 2 Abstract The goal of this study is to evolve novel seeds for Conway s Game

More information

Artificial Neural Networks Examination, March 2004

Artificial Neural Networks Examination, March 2004 Artificial Neural Networks Examination, March 2004 Instructions There are SIXTY questions (worth up to 60 marks). The exam mark (maximum 60) will be added to the mark obtained in the laborations (maximum

More information

RNA evolution and Genotype to Phenotype maps

RNA evolution and Genotype to Phenotype maps RNA evolution and Genotype to Phenotype maps E.S. Colizzi November 8, 2018 Introduction Biological evolution occurs in a population because 1) different genomes can generate different reproductive success

More information

Evolutionary computation

Evolutionary computation Evolutionary computation Andrea Roli andrea.roli@unibo.it Dept. of Computer Science and Engineering (DISI) Campus of Cesena Alma Mater Studiorum Università di Bologna Outline 1 Basic principles 2 Genetic

More information

Achieving High-Level Functionality through Complexification

Achieving High-Level Functionality through Complexification From: AAAI Technical Report SS-03-02. Compilation copyright 2003, AAAI (www.aaai.org). All rights reserved. Achieving High-Level Functionality through Complexification Kenneth O. Stanley and Risto Miikkulainen

More information

Lecture 9 Evolutionary Computation: Genetic algorithms

Lecture 9 Evolutionary Computation: Genetic algorithms Lecture 9 Evolutionary Computation: Genetic algorithms Introduction, or can evolution be intelligent? Simulation of natural evolution Genetic algorithms Case study: maintenance scheduling with genetic

More information

Phylogeny and systematics. Why are these disciplines important in evolutionary biology and how are they related to each other?

Phylogeny and systematics. Why are these disciplines important in evolutionary biology and how are they related to each other? Phylogeny and systematics Why are these disciplines important in evolutionary biology and how are they related to each other? Phylogeny and systematics Phylogeny: the evolutionary history of a species

More information

Environmental Influence on the Evolution of Morphological Complexity in Machines

Environmental Influence on the Evolution of Morphological Complexity in Machines Environmental Influence on the Evolution of Morphological Complexity in Machines Joshua E. Auerbach 1 *, Josh C. Bongard 2 1 Laboratory of Intelligent Systems, École Polytechnique Fédérale de Lausanne

More information

Neuroevolution for Reinforcement Learning Using Evolution Strategies

Neuroevolution for Reinforcement Learning Using Evolution Strategies Neuroevolution for Reinforcement Learning Using Evolution Strategies Christian Igel Institut für Neuroinformatik Ruhr-Universität Bochum 44780 Bochum, Germany christian.igel@neuroinformatik.rub.de Abstract-

More information

Using Embryonic Stages to increase the evolvability of development

Using Embryonic Stages to increase the evolvability of development Using Embryonic Stages to increase the evolvability of development Diego Federici Norwegian University of Science and Technology Department of computer and information science N-749 Trondheim, Norway federici@idi.ntnu.no

More information

0 o 1 i B C D 0/1 0/ /1

0 o 1 i B C D 0/1 0/ /1 A Comparison of Dominance Mechanisms and Simple Mutation on Non-Stationary Problems Jonathan Lewis,? Emma Hart, Graeme Ritchie Department of Articial Intelligence, University of Edinburgh, Edinburgh EH

More information

Formalizing the gene centered view of evolution

Formalizing the gene centered view of evolution Chapter 1 Formalizing the gene centered view of evolution Yaneer Bar-Yam and Hiroki Sayama New England Complex Systems Institute 24 Mt. Auburn St., Cambridge, MA 02138, USA yaneer@necsi.org / sayama@necsi.org

More information

[Read Chapter 9] [Exercises 9.1, 9.2, 9.3, 9.4]

[Read Chapter 9] [Exercises 9.1, 9.2, 9.3, 9.4] 1 EVOLUTIONARY ALGORITHMS [Read Chapter 9] [Exercises 9.1, 9.2, 9.3, 9.4] Evolutionary computation Prototypical GA An example: GABIL Schema theorem Genetic Programming Individual learning and population

More information

Neural Systems and Artificial Life Group, Institute of Psychology, National Research Council, Rome. Evolving Modular Architectures for Neural Networks

Neural Systems and Artificial Life Group, Institute of Psychology, National Research Council, Rome. Evolving Modular Architectures for Neural Networks Neural Systems and Artificial Life Group, Institute of Psychology, National Research Council, Rome Evolving Modular Architectures for Neural Networks Andrea Di Ferdinando, Raffaele Calabretta and Domenico

More information

What does it take to evolve behaviorally complex organisms?

What does it take to evolve behaviorally complex organisms? 1 Neural Systems and Artificial Life Group Institute of Psychology National Research Council, Rome What does it take to evolve behaviorally complex organisms? Raffaele Calabretta 1, Andrea Di Ferdinando

More information

Chapter 5 Evolution of Biodiversity

Chapter 5 Evolution of Biodiversity Chapter 5 Evolution of Biodiversity Earth is home to a tremendous diversity of species diversity- the variety of ecosystems within a given region. diversity- the variety of species in a given ecosystem.

More information

The Relationship Between Evolvability and Robustness in the Evolution of Boolean Networks

The Relationship Between Evolvability and Robustness in the Evolution of Boolean Networks The Relationship Between Evolvability and Robustness in the Evolution of Boolean Networks David Shorten 1 and Geoff Nitschke 1 1 Department of Computer Science, University of Cape Town dshorten@cs.uct.ac.za,

More information

Development. biologically-inspired computing. lecture 16. Informatics luis rocha x x x. Syntactic Operations. biologically Inspired computing

Development. biologically-inspired computing. lecture 16. Informatics luis rocha x x x. Syntactic Operations. biologically Inspired computing lecture 16 -inspired S S2 n p!!! 1 S Syntactic Operations al Code:N Development x x x 1 2 n p S Sections I485/H400 course outlook Assignments: 35% Students will complete 4/5 assignments based on algorithms

More information

Neuroevolution: from architectures to learning

Neuroevolution: from architectures to learning Evol. Intel. (2008) 1:47 62 DOI 10.1007/s12065-007-0002-4 REVIEW ARTICLE Neuroevolution: from architectures to learning Dario Floreano Æ Peter Dürr Æ Claudio Mattiussi Received: 5 October 2007 / Accepted:

More information

Evolution Unit: What is Evolution?

Evolution Unit: What is Evolution? Evolution Unit: What is Evolution? What is The Theory of Evolution? Evolution is, a change (in the genetic composition) of a population over time. on a larger scale, the entire biological history, from

More information

Haploid & diploid recombination and their evolutionary impact

Haploid & diploid recombination and their evolutionary impact Haploid & diploid recombination and their evolutionary impact W. Garrett Mitchener College of Charleston Mathematics Department MitchenerG@cofc.edu http://mitchenerg.people.cofc.edu Introduction The basis

More information

Genetic assimilation can occur in the absence of selection for the assimilating phenotype, suggesting a role for the canalization heuristic

Genetic assimilation can occur in the absence of selection for the assimilating phenotype, suggesting a role for the canalization heuristic doi: 10.1111/j.1420-9101.2004.00739.x Genetic assimilation can occur in the absence of selection for the assimilating phenotype, suggesting a role for the canalization heuristic J. MASEL Department of

More information

HyperNEAT for Locomotion Control in Modular Robots

HyperNEAT for Locomotion Control in Modular Robots Published in Evolvable Systems: From Biology to Hardware 2010; Gianluca Tempesti, Andy Tyrrell and Julian Miller, editors. The original publication is available at www.springerlink. com. HyperNEAT for

More information

Measuring and Optimizing Behavioral Complexity for Evolutionary Reinforcement Learning

Measuring and Optimizing Behavioral Complexity for Evolutionary Reinforcement Learning Measuring and Optimizing Behavioral Complexity for Evolutionary Reinforcement Learning Faustino J. Gomez, Julian Togelius, and Juergen Schmidhuber IDSIA, Galleria 2 6928 Manno-Lugano Switzerland Abstract.

More information

Using NEAT to Stabilize an Inverted Pendulum

Using NEAT to Stabilize an Inverted Pendulum Using NEAT to Stabilize an Inverted Pendulum Awjin Ahn and Caleb Cochrane May 9, 2014 Abstract The inverted pendulum balancing problem is a classic benchmark problem on which many types of control implementations

More information

Genetics and Natural Selection

Genetics and Natural Selection Genetics and Natural Selection Darwin did not have an understanding of the mechanisms of inheritance and thus did not understand how natural selection would alter the patterns of inheritance in a population.

More information

NCEA Level 2 Biology (91157) 2017 page 1 of 5 Assessment Schedule 2017 Biology: Demonstrate understanding of genetic variation and change (91157)

NCEA Level 2 Biology (91157) 2017 page 1 of 5 Assessment Schedule 2017 Biology: Demonstrate understanding of genetic variation and change (91157) NCEA Level 2 Biology (91157) 2017 page 1 of 5 Assessment Schedule 2017 Biology: Demonstrate understanding of genetic variation and change (91157) Evidence Statement Q1 Expected coverage Merit Excellence

More information

Evolutionary Algorithms

Evolutionary Algorithms Evolutionary Algorithms a short introduction Giuseppe Narzisi Courant Institute of Mathematical Sciences New York University 31 January 2008 Outline 1 Evolution 2 Evolutionary Computation 3 Evolutionary

More information

Evolution of Neural Networks in a Multi-Agent World

Evolution of Neural Networks in a Multi-Agent World ZESZYTY NAUKOWE UNIWERSYTETU JAGIELLOŃSKIEGO MMCCXLII PRACE INFORMATYCZNE Z. 10 2000 Evolution of Neural Networks in a Multi-Agent World Marek Kisiel-Dorohinicki Magdalena Klapper-Rybicka Institute of

More information

arxiv: v1 [cs.ne] 5 Feb 2013

arxiv: v1 [cs.ne] 5 Feb 2013 1 Evolvability Is Inevitable: Increasing Evolvability Without the Pressure to Adapt Joel Lehman 1,, Kenneth O. Stanley 2 1 Department of Computer Science, The University of Texas at Austin, Austin, TX,

More information

Evolution of Bilateral Symmetry in Agents Controlled by Spiking Neural Networks

Evolution of Bilateral Symmetry in Agents Controlled by Spiking Neural Networks Evolution of Bilateral Symmetry in Agents Controlled by Spiking Neural Networks Nicolas Oros, Volker Steuber, Neil Davey, Lola Cañamero, and Rod Adams Science and Technology Research Institute, University

More information

BIOLOGY. Monday 2 May 2016

BIOLOGY. Monday 2 May 2016 BIOLOGY Monday 2 May 2016 Entry Task What does it mean when a populations allele frequency changes from 20% to 30%? The population is evolving. Agenda Entry Task Housekeeping Exam Study (Kahoot) Housekeeping

More information

The Common Ground Curriculum. Science: Biology

The Common Ground Curriculum. Science: Biology The Common Ground Curriculum Science: Biology CGC Science : Biology Defining Biology: Biology is the study of living things in their environment. This is not a static, snapshot of the living world but

More information

Reproduction and Evolution Practice Exam

Reproduction and Evolution Practice Exam Reproduction and Evolution Practice Exam Topics: Genetic concepts from the lecture notes including; o Mitosis and Meiosis, Homologous Chromosomes, Haploid vs Diploid cells Reproductive Strategies Heaviest

More information

Integer weight training by differential evolution algorithms

Integer weight training by differential evolution algorithms Integer weight training by differential evolution algorithms V.P. Plagianakos, D.G. Sotiropoulos, and M.N. Vrahatis University of Patras, Department of Mathematics, GR-265 00, Patras, Greece. e-mail: vpp

More information

Curriculum Links. AQA GCE Biology. AS level

Curriculum Links. AQA GCE Biology. AS level Curriculum Links AQA GCE Biology Unit 2 BIOL2 The variety of living organisms 3.2.1 Living organisms vary and this variation is influenced by genetic and environmental factors Causes of variation 3.2.2

More information

Encouraging Behavioral Diversity in Evolutionary Robotics: an Empirical Study

Encouraging Behavioral Diversity in Evolutionary Robotics: an Empirical Study Encouraging Behavioral Diversity in Evolutionary Robotics: an Empirical Study J.-B. Mouret jean-baptiste.mouret@upmc.fr ISIR, Université Pierre et Marie Curie-Paris 6, CNRS UMR 7222, Paris, F-75252, Paris

More information

On the Difficulty of Co-Optimizing Morphology and Control in Evolved Virtual Creatures

On the Difficulty of Co-Optimizing Morphology and Control in Evolved Virtual Creatures On the Difficulty of Co-Optimizing Morphology and Control in Evolved Virtual Creatures Nick Cheney 1, Josh Bongard 2, Vytas SunSpiral 3 and Hod Lipson 4 1 Department of Biological Statistics and Computational

More information

Evolving Neural Networks in Compressed Weight Space

Evolving Neural Networks in Compressed Weight Space Evolving Neural Networks in Compressed Weight Space Jan Koutník IDSIA University of Lugano Manno-Lugano, CH hkou@idsia.ch Faustino Gomez IDSIA University of Lugano Manno-Lugano, CH tino@idsia.ch Jürgen

More information

A Genetic Algorithm and an Exact Algorithm for Classifying the Items of a Questionnaire Into Different Competences

A Genetic Algorithm and an Exact Algorithm for Classifying the Items of a Questionnaire Into Different Competences A Genetic Algorithm and an Exact Algorithm for Classifying the Items of a Questionnaire Into Different Competences José Luis Galán-García 1, Salvador Merino 1 Javier Martínez 1, Miguel de Aguilera 2 1

More information

Behavioral Diversity with Multiple Behavioral Distances

Behavioral Diversity with Multiple Behavioral Distances Doncieux, S. and Mouret, J.-B. (2013), Behavioral Diversity with Multiple Behavioral Distances, IEEE Congress on Evolutionary Computation, 2013 (CEC 2013). Behavioral Diversity with Multiple Behavioral

More information

On Computational Limitations of Neural Network Architectures

On Computational Limitations of Neural Network Architectures On Computational Limitations of Neural Network Architectures Achim Hoffmann + 1 In short A powerful method for analyzing the computational abilities of neural networks based on algorithmic information

More information

MATHEMATICAL MODELS - Vol. III - Mathematical Modeling and the Human Genome - Hilary S. Booth MATHEMATICAL MODELING AND THE HUMAN GENOME

MATHEMATICAL MODELS - Vol. III - Mathematical Modeling and the Human Genome - Hilary S. Booth MATHEMATICAL MODELING AND THE HUMAN GENOME MATHEMATICAL MODELING AND THE HUMAN GENOME Hilary S. Booth Australian National University, Australia Keywords: Human genome, DNA, bioinformatics, sequence analysis, evolution. Contents 1. Introduction:

More information

Principles of Modularity, Regularity, and Hierarchy for Scalable Systems

Principles of Modularity, Regularity, and Hierarchy for Scalable Systems Principles of Modularity, Regularity, and Hierarchy for Scalable Systems Hod Lipson Mechanical & Aerospace Engineering, and Computing & Information Science, Cornell University, Ithaca NY 14850, USA hod.lipson@cornell.edu

More information

Lecture 15: Genetic Algorithms

Lecture 15: Genetic Algorithms Lecture 15: Genetic Algorithms Dr Roman V Belavkin BIS3226 Contents 1 Combinatorial Problems 1 2 Natural Selection 2 3 Genetic Algorithms 3 31 Individuals and Population 3 32 Fitness Functions 3 33 Encoding

More information

IAC-16-D1.2.1 (34366) Automated Design of CubeSats and Small Spacecrafts

IAC-16-D1.2.1 (34366) Automated Design of CubeSats and Small Spacecrafts IAC-16-D1.2.1 (34366) Automated Design of CubeSats and Small Spacecrafts Himangshu Kalita a, Jekanthan Thangavelautham b* a School of Energy, Matter and Transport Engineering, Arizona State University,

More information

Inferring a System of Differential Equations for a Gene Regulatory Network by using Genetic Programming

Inferring a System of Differential Equations for a Gene Regulatory Network by using Genetic Programming Inferring a System of Differential Equations for a Gene Regulatory Network by using Genetic Programming Erina Sakamoto School of Engineering, Dept. of Inf. Comm.Engineering, Univ. of Tokyo 7-3-1 Hongo,

More information

How should we organize the diversity of animal life?

How should we organize the diversity of animal life? How should we organize the diversity of animal life? The difference between Taxonomy Linneaus, and Cladistics Darwin What are phylogenies? How do we read them? How do we estimate them? Classification (Taxonomy)

More information

Lecture 22. Introduction to Genetic Algorithms

Lecture 22. Introduction to Genetic Algorithms Lecture 22 Introduction to Genetic Algorithms Thursday 14 November 2002 William H. Hsu, KSU http://www.kddresearch.org http://www.cis.ksu.edu/~bhsu Readings: Sections 9.1-9.4, Mitchell Chapter 1, Sections

More information

Evolution and Development of a Multi-Cellular Organism: Scalability, Resilience and Neutral. complexification

Evolution and Development of a Multi-Cellular Organism: Scalability, Resilience and Neutral. complexification Evolution and Development of a Multi-Cellular Organism: Scalability, Resilience and Neutral Complexification Diego Federici and Keith Downing Norwegian University of Science and Technology Department of

More information

Evolutionary Computation. DEIS-Cesena Alma Mater Studiorum Università di Bologna Cesena (Italia)

Evolutionary Computation. DEIS-Cesena Alma Mater Studiorum Università di Bologna Cesena (Italia) Evolutionary Computation DEIS-Cesena Alma Mater Studiorum Università di Bologna Cesena (Italia) andrea.roli@unibo.it Evolutionary Computation Inspiring principle: theory of natural selection Species face

More information

The genomes of recombinant inbred lines

The genomes of recombinant inbred lines The genomes of recombinant inbred lines Karl W Broman Department of Biostatistics Johns Hopkins University http://www.biostat.jhsph.edu/~kbroman C57BL/6 2 1 Recombinant inbred lines (by sibling mating)

More information

I N N O V A T I O N L E C T U R E S (I N N O l E C) Petr Kuzmič, Ph.D. BioKin, Ltd. WATERTOWN, MASSACHUSETTS, U.S.A.

I N N O V A T I O N L E C T U R E S (I N N O l E C) Petr Kuzmič, Ph.D. BioKin, Ltd. WATERTOWN, MASSACHUSETTS, U.S.A. I N N O V A T I O N L E C T U R E S (I N N O l E C) Binding and Kinetics for Experimental Biologists Lecture 2 Evolutionary Computing: Initial Estimate Problem Petr Kuzmič, Ph.D. BioKin, Ltd. WATERTOWN,

More information

Introduction to population genetics & evolution

Introduction to population genetics & evolution Introduction to population genetics & evolution Course Organization Exam dates: Feb 19 March 1st Has everybody registered? Did you get the email with the exam schedule Summer seminar: Hot topics in Bioinformatics

More information

Artificial Neural Networks Examination, June 2005

Artificial Neural Networks Examination, June 2005 Artificial Neural Networks Examination, June 2005 Instructions There are SIXTY questions. (The pass mark is 30 out of 60). For each question, please select a maximum of ONE of the given answers (either

More information

Molecular Evolution & the Origin of Variation

Molecular Evolution & the Origin of Variation Molecular Evolution & the Origin of Variation What Is Molecular Evolution? Molecular evolution differs from phenotypic evolution in that mutations and genetic drift are much more important determinants

More information

Molecular Evolution & the Origin of Variation

Molecular Evolution & the Origin of Variation Molecular Evolution & the Origin of Variation What Is Molecular Evolution? Molecular evolution differs from phenotypic evolution in that mutations and genetic drift are much more important determinants

More information

Processes of Evolution

Processes of Evolution 15 Processes of Evolution Forces of Evolution Concept 15.4 Selection Can Be Stabilizing, Directional, or Disruptive Natural selection can act on quantitative traits in three ways: Stabilizing selection

More information

Fundamentals of Genetic Algorithms

Fundamentals of Genetic Algorithms Fundamentals of Genetic Algorithms : AI Course Lecture 39 40, notes, slides www.myreaders.info/, RC Chakraborty, e-mail rcchak@gmail.com, June 01, 2010 www.myreaders.info/html/artificial_intelligence.html

More information

An Improved System for Artificial Creatures Evolution

An Improved System for Artificial Creatures Evolution An Improved System for Artificial Creatures Evolution Thomas Miconi 1 and Alastair Channon 1 1 University of Birmingham, Edgbaston B152TT, Birmingham, UK t.miconi@cs.bham.ac.uk Abstract We present our

More information