Evolving Multimodal Networks for Multitask Games

Slides:



Advertisements
Similar presentations
Non-dominated Sorting Genetic Algorithm (NSGA-II)
Advertisements

MOEAs University of Missouri - Rolla Dr. T’s Course in Evolutionary Computation Matt D. Johnson November 6, 2006.
Yuri R. Tsoy, Vladimir G. Spitsyn, Department of Computer Engineering
Constructing Complex NPC Behavior via Multi- Objective Neuroevolution Jacob Schrum – Risto Miikkulainen –
Elitist Non-dominated Sorting Genetic Algorithm: NSGA-II
A Study on Recent Fast Ways of Hypervolume Calculation for MOEAs Mainul Kabir ( ) and Nasik Muhammad Nafi ( ) Department of Computer Science.
1 Part I Artificial Neural Networks Sofia Nikitaki.
Estimation of Distribution Algorithms Ata Kaban School of Computer Science The University of Birmingham.
Multi-Objective Evolutionary Algorithms Matt D. Johnson April 19, 2007.
Evolving Neural Network Agents in the NERO Video Game Author : Kenneth O. Stanley, Bobby D. Bryant, and Risto Miikkulainen Presented by Yi Cheng Lin.
Using a GA to Create Prey Tactics Presented by Tony Morelli on 11/29/04.
Genetic Algorithms in Materials Processing N. Chakraborti Department of Metallurgical & Materials Engineering Indian Institute of Technology Kharagpur.
Design of Autonomous Navigation Controllers for Unmanned Aerial Vehicles using Multi-objective Genetic Programming Gregory J. Barlow March 19, 2004.
Torcs Simulator Presented by Galina Volkinshtein and Evgenia Dubrovsky.
1 Autonomous Controller Design for Unmanned Aerial Vehicles using Multi-objective Genetic Programming Gregory J. Barlow North Carolina State University.
Design of Curves and Surfaces by Multi Objective Optimization Rony Goldenthal Michel Bercovier School of Computer Science and Engineering The Hebrew University.
Othello Sean Farrell June 29, Othello Two-player game played on 8x8 board All pieces have one white side and one black side Initial board setup.
Marcus Gallagher and Mark Ledwich School of Information Technology and Electrical Engineering University of Queensland, Australia Sumaira Saeed Evolving.
By Jacob Schrum and Risto Miikkulainen
Evolving Multi-modal Behavior in NPCs Jacob Schrum – Risto Miikkulainen –
Evolution, Brains and Multiple Objectives
On comparison of different approaches to the stability radius calculation Olga Karelkina Department of Mathematics University of Turku MCDM 2011.
Evolving a Sigma-Pi Network as a Network Simulator by Justin Basilico.
Cristian Urs and Ben Riveira. Introduction The article we chose focuses on improving the performance of Genetic Algorithms by: Use of predictive models.
Soft Computing Lecture 18 Foundations of genetic algorithms (GA). Using of GA.
MOGADES: Multi-Objective Genetic Algorithm with Distributed Environment Scheme Intelligent Systems Design Laboratory , Doshisha University , Kyoto Japan.
Constructing Intelligent Agents via Neuroevolution By Jacob Schrum
Efficient and Scalable Computation of the Energy and Makespan Pareto Front for Heterogeneous Computing Systems Kyle M. Tarplee 1, Ryan Friese 1, Anthony.
Study on Genetic Network Programming (GNP) with Learning and Evolution Hirasawa laboratory, Artificial Intelligence section Information architecture field.
More on coevolution and learning Jing Xiao April, 2008.
Evolution of Teamwork in Multiagent Systems Research Preparation Examination by Jacob Schrum.
Pareto Coevolution Presented by KC Tsui Based on [1]
Omni-Optimizer A Procedure for Single and Multi-objective Optimization Prof. Kalyanmoy Deb and Santosh Tiwari.
Genetic Algorithms Genetic Algorithms – What are they? And how they are inspired from evolution. Operators and Definitions in Genetic Algorithms paradigm.
CAP6938 Neuroevolution and Developmental Encoding Real-time NEAT Dr. Kenneth Stanley October 18, 2006.
Mike Taks Bram van de Klundert. About Published 2005 Cited 286 times Kenneth O. Stanley Associate Professor at University of Central Florida Risto Miikkulainen.
ANNIE EA Papers Sampling the Nature of a Population: Punctuated Anytime Learning for Co- Evolving a Team – Gary Parker, H. Joseph Blumenthal Ants and Evolution:
Kanpur Genetic Algorithms Laboratory IIT Kanpur 25, July 2006 (11:00 AM) Multi-Objective Dynamic Optimization using Evolutionary Algorithms by Udaya Bhaskara.
DIVERSITY PRESERVING EVOLUTIONARY MULTI-OBJECTIVE SEARCH Brian Piper1, Hana Chmielewski2, Ranji Ranjithan1,2 1Operations Research 2Civil Engineering.
Chapter 9 Genetic Algorithms.  Based upon biological evolution  Generate successor hypothesis based upon repeated mutations  Acts as a randomized parallel.
Advanced Computer Architecture & Processing Systems Research Lab Framework for Automatic Design Space Exploration.
Pac-Man AI using GA. Why Machine Learning in Video Games? Better player experience Agents can adapt to player Increased variety of agent behaviors Ever-changing.
2/29/20121 Optimizing LCLS2 taper profile with genetic algorithms: preliminary results X. Huang, J. Wu, T. Raubenhaimer, Y. Jiao, S. Spampinati, A. Mandlekar,
Artificial Intelligence Research in Video Games By Jacob Schrum
CAP6938 Neuroevolution and Artificial Embryogeny Competitive Coevolution Dr. Kenneth Stanley February 20, 2006.
Evolving Agent Behavior in Multiobjective Domains Using Fitness-Based Shaping Jacob Schrum and Risto Miikkulainen University of Texas at Austin Department.
Iterated Prisoner’s Dilemma Game in Evolutionary Computation Seung-Ryong Yang.
Comparative Reproduction Schemes for Evolving Gathering Collectives A.E. Eiben, G.S. Nitschke, M.C. Schut Computational Intelligence Group Department of.
CAP6938 Neuroevolution and Developmental Encoding Intro to Neuroevolution Dr. Kenneth Stanley September 18, 2006.
Multi-objective Evolutionary Algorithms (for NACST/Seq) summarized by Shin, Soo-Yong.
Riza Erdem Jappie Klooster Dirk Meulenbelt EVOLVING MULTI-MODAL BEHAVIOR IN NPC S.
Competition and Cooperation

Zhengli Huang and Wenliang (Kevin) Du
Botprize 2010 Jacob Schrum, Igor Karpov, and Risto Miikkulainen
CAP6938 Neuroevolution and Artificial Embryogeny Real-time NEAT Dr. Kenneth Stanley February 22, 2006.
An Evolutionary Algorithm for Neural Network Learning using Direct Encoding Paul Batchis Department of Computer Science Rutgers University.
Evolving robot brains using vision Lisa Meeden Computer Science Department Swarthmore College.
Genetic Algorithm(GA)
UT^2: Human-like Behavior via Neuroevolution of Combat Behavior and Replay of Human Traces Jacob Schrum Igor V. Karpov
Solving Interleaved and Blended Sequential Decision-Making Problems through Modular Neuroevolution Jacob Schrum Risto Miikkulainen.
TORCS WORKS Jang Su-Hyung.
Evolutionary Algorithms Jim Whitehead
Dr. Kenneth Stanley January 30, 2006
Presented by: Dr Beatriz de la Iglesia
Dr. Kenneth Stanley September 25, 2006
Evolving Multimodal Networks for Multitask Games
UT^2: Human-like Behavior via Neuroevolution of Combat Behavior and Replay of Human Traces Jacob Schrum, Igor Karpov, and Risto Miikkulainen
Dr. Kenneth Stanley February 6, 2006
Lecture 4. Niching and Speciation (1)
Presentation transcript:

Evolving Multimodal Networks for Multitask Games Jacob Schrum – schrum2@cs.utexas.edu Risto Miikkulainen – risto@cs.utexas.edu University of Texas at Austin Department of Computer Science

Evolution in videogames Automatically learn interesting behavior Complex but controlled environments Stepping stone to real world Robots Training simulators Complexity issues Multiple contradictory objectives Multiple challenging tasks

Multitask Games NPCs perform two or more separate tasks Each task has own performance measures Task linkage Independent Dependent Not blended Inherently multiobjective

Test Domains Designed to study multimodal behavior Two tasks in similar environments Different behavior needed to succeed Main challenge: perform well in both Front Ramming Back Ramming

Front/Back Ramming Same goal, opposite embodiments Front Ramming Attack w/front ram Avoid counterattacks Back Ramming Attack w/back ram Avoid counterattacks

Predator/Prey Same embodiment, opposite goals Predator Prey Attack prey Prevent escape Prey Avoid attack Stay alive

Multiobjective Optimization High health but did not deal much damage Game with two objectives: Damage Dealt Remaining Health A dominates B iff A is strictly better in one objective and at least as good in others Population of points not dominated are best: Pareto Front Weighted-sum provably incapable of capturing non-convex front Tradeoff between objectives Dealt lot of damage, but lost lots of health

NSGA-II Evolution: natural approach for finding optimal population Non-Dominated Sorting Genetic Algorithm II* Population P with size N; Evaluate P Use mutation to get P´ size N; Evaluate P´ Calculate non-dominated fronts of {P È P´} size 2N New population size N from highest fronts of {P È P´} *K. Deb et al. A Fast and Elitist Multiobjective Genetic Algorithm: NSGA-II. Evol. Comp. 2002

Constructive Neuroevolution Genetic Algorithms + Neural Networks Build structure incrementally (complexification) Good at generating control policies Three basic mutations (no crossover used) Perturb Weight Add Connection Add Node

Multimodal Networks (1) Multitask Learning* One mode per task Shared hidden layer Knows current task Previous work Supervised learning context Multiple tasks learned quicker than individual Not tried with evolution yet * R. A. Caruana, "Multitask learning: A knowledge-based source of inductive bias" ICML 1993

Multimodal Networks (2) Starting network with one mode Mode Mutation Extra modes evolved Networks choose mode Chosen via preference neurons MM Previous Links from previous mode Weights = 1.0 MM Random Links from random sources Random weights Supports mode deletion MM(P) MM(R)

Experiment Compare 4 conditions: 500 generations Population size 52 Control: Unimodal networks Multitask: One mode per task MM(P): Mode Mutation Previous MM(R): Mode Mutation Random + Delete Mutation 500 generations Population size 52 “Player” behavior scripted Network controls homogeneous team of 4

MO Performance Assessment Reduce Pareto front to single number Hypervolume of dominated region Pareto compliant Front A dominates front B implies HV(A) > HV(B) Standard statistical comparisons of average HV

20 runs

Front/Back Ramming Behaviors Multitask Front Ramming Back Ramming MM(R)

20 runs

Predator/Prey Behaviors Multitask Prey Predator MM(R)

Discussion (1) Front/Back Ramming Control < MM(P), MM(R) < Multitask Multiple modes help Explicit knowledge of task helps

Discussion (2) Predator/Prey MM(P), Control, Multitask < MM(R) Multiple modes not necessarily helpful Disparity in relative difficulty of tasks Multitask ends up wasting effort Mode deletion aids search for one good mode

How To Apply Multitask good if: Mode mutation good if: Task division known, and Tasks are comparably difficult Mode mutation good if: Task division is unknown, or “Obvious” task division is misleading

Future Work Games with more tasks Games with independent tasks Does method scale? Control mode bloat Games with independent tasks Ms. Pac-Man Collect pills while avoiding ghosts Eat ghosts after eating power pill Games with blended tasks Unreal Tournament 2004 Fight while avoiding damage Fight or run away? Collect items or seek opponents?

Conclusion Domains with multiple tasks are common Both in real world and games Multimodal networks improve learning in multitask games Will allow interesting/complex behavior to be developed in future

Questions? Jacob Schrum – schrum2@cs.utexas.edu Risto Miikkulainen – risto@cs.utexas.edu University of Texas at Austin Department of Computer Science

Auxiliary Slides