Kenneth O. Stanley

Assistant Professor

University of Central Florida School of Electrical Engineering and Computer Science


kstanley@cs.ucf.edu
Office: Harris 332

Vita: PDF

Quick Links: Meeting | Teaching | Research | Service | NEAT Users | Software/Source Code | Animated Demos | Publications |

EPlex Research Group

NERO Project | Dance Evolution (new)

Research

I am the director of the Evolutionary Complexity (EPlex) Research Group at UCF. Our research focuses on abstracting the essential properties of natural evolution that made it possible to discover astronomically complex structures such as the human brain.

I developed a method, called NEAT (NeuroEvolution of Augmenting Topologies), that begins evolution with a population of very simple networks and complexifies the networks over generations by adding new neurons and connections.

We recently developed an extension to NEAT called HyperNEAT that can evolve neural networks with millions of connections and exploit geometric regularities in the task domain.. Publications on HyperNEAT include HyperNEAT applied to checkers (AAAI-08), HyperNEAT in multiagent learning (GECCO-08), HyperNEAT applied to a simple vision task with massive scaling (GECCO-07), and HyperNEAT in robot control with differing geometries (GECCO-07). HyperNEAT source code is also available in C# and C++.

I discuss some of my research interests in this interview conducted by Tom Barbalet for biota.org.

Before joining UCF, I was a member of the UTCS Neural Networks Research Group.

Günter Bachelier created a gallery of computerized portraits of myself , generated through an evolutionary process, as part of our recognition for winning the Best Paper Award at EvoMUSART 2008. Gunter added another set here

Service Activities and Invited Talks

I delivered an invited plenary lecture on HyperNEAT at the 18th International Conference on Artificial Neural Networks (ICANN 2008).

I am the chair of the IEEE Task Force on Computational Intelligence in Video Games. If you are interested in the combination of AI and video games, please feel free to join the task force by sending me an email

In 2008, I co-chair with Sanjeev Kumar the Generative Developmental Systems track at GECCO-2008. We were founding co-chairs of the track, which is the only exclusive conference venue for papers in GDS, along with Julian Miller in 2007.

I am on the program committee of the AAAI-08 AI Video Competition. Accepted videos will be showcased at AAAI-08 in Chicago, including Shakey awards for the best videos and $3,000 in prizes. Follow the link for submission instructions!

Teaching

Current Semester: Fall 2008: NeuroEvolution and Generative and Developmental Systems (CAP 6616): Syllabus | Lectures and Assignments

Old: Spring 2008: AI for Game Programming (CAP 4932): Syllabus | Lectures and Assignments

Old: Fall 2007: NeuroEvolution and Generative and Developmental Systems (CAP 6616): Syllabus | Lectures and Assignments

Old: Spring 2007: Machine Learning (CAP 5810): Syllabus | Lectures and Assignments

Old: Fall 2006: Topics in NeuroEvolution and Developmental Encoding (CAP 6938): Syllabus | Lectures and Assignments

Machine Learning I (EEL 4818; Undergraduate Honors) Class Homepage (Co-taught with Drs. R. DeMara, M. Georgiopoulos, M. Mollaghasemi, and I. Garibay)

Old: Spring 2006: Topics in NeuroEvolution (CAP 6938)

NEAT Users

Derek James runs a NEAT Users Group on Yahoo. Please feel free to join to discuss NEAT-related issues.

If you are a NEAT user, or interested in working with NEAT, please see the NEAT Users Page , which includes a helpful FAQ.

Schedule Meeting

You can schedule a meeting with me here: Online Meeting Schedule (please follow the instructions there)

Note that you must be logged into the EPlex website in order to use this feature. If you see the message, "You are not authorized to view this resource," then you need to log in first and the page will appear. If you do not have a login, please send me an email to schedule a meeting.

Picbreeder

(click logo to enter)

Picbreeder is the world's largest collaborative interactive evolution online service.

Here is a sample of evolutionary art from Picbreeder

Dance Evolution

This undergraduate research project created with Jeff Balogh, Greg Dubbin, and Michael Do is a new entertainment application that allows users to interactively evolve dancers that make up their own moves to any MIDI song.

Dance Evolution has been released. It is available here.

We recently submitted a video from our project to the first ever AAAI-07 AI Video Competion.
It won the Best Student Video Award on 7/23/07.
(It was also a nominee for Best Explanation.)

On the left is a slightly muddy Youtube version of the video and on the right is a single dance clip from me recently playing with a later version of the program on my own:

We plan to release Dance Evolution for public download at a future date.

NERO

NERO (NeuroEvolving Robotic Operatives) is a new kind of video game that utilizes rtNEAT (real-time NEAT) at its core technology.

Software / Source Code

Dance Evolution

Software and source code for both Windows and Linux are available here. Dance evolution allows users to train 3D animated characters to dance to any MIDI song using interactive evolution.

Novelty Search

Novelty Search C++ by Joel Lehman implements the novelty search algorithm introduced by Lehman and Stanley in Exploiting Open-Endedness to Solve Problems Through the Search for Novelty . Experiments in the paper are included in the software release.

HyperNEAT Versions

HyperNEAT is Hypercube-based NeuroEvolution of Augmenting Topologies, a new algorithm for evolving very large neural networks with geometric regularities. More information is available under publications.

NEAT Versions

NEAT-based Genetic Art (CPPN Explorer) Programs

For an introduction to the theory behind Compositional Pattern Producing Networks (CPPNs), which explains how the patterns produced by these programs are possible to evolve, please refer to our paper Exploiting Regularity Without Development.

Animated Demos

Animations of evolved NEAT neural network behavior: Robot Duel, hall navigation, function approximation, NEAT Hopper, and Peons. Demo formats include GIF animations, .avi movies, and a Windows executable.

The Evolution of a Spaceship

Watch how NEAT neural networks evolved to draw a multi-featured spaceship out of scratch. A fascinating exhibition of complexification and the evolution of form.

Paperback Book Available with Chapter on NEAT

For people who are interested in learning about NEAT, but prefer explanations intended for general audiences to reading research-level papers, I am happy to recommend AI Techniques for Game Programming by Mat Buckland. Most of the final chapter of this book describes NEAT in a fun and simple style. The book also comes with source code. This book is a good resource for hobbyists or video game programmers interested in AI techniques. (Researchers should still refer to the NEAT research publications available below.) It also includes useful introductions to genetic algorithms and neural networks.
To those who already own the book: Feel free to email me questions or ideas you have regarding NEAT. I am happy to receive feedback.

Selected Publications

(For a full list of my publications, including links, please see my vita)

Dissertation

Ph.D. Dissertation: EFFICIENT EVOLUTION OF NEURAL NETWORKS THROUGH COMPLEXIFICATION
Kenneth O. Stanley
Department of Computer Sciences, The University of Texas at Austin
Technical Report~AI-TR-04-39, August 2004.

Journal Papers

A Hypercube-Based Indirect Encoding for Evolving Large-Scale Neural Networks
Kenneth O. Stanley, David B. D'Ambrosio, and Jason Gauci
School of Electrical Engineering and Computer Science, University of Central Florida
To appear in: Artificial Life journal. Cambridge, MA: MIT Press, 2009. Manuscript 39 pages

Picbreeder: Collaborative Interactive Evolution of Images
Jimmy Secretan, Nicholas Beato, David B. D'Ambrosio, Adelein Rodriguez, Adam Campbell, and Kenneth O. Stanley
School of Electrical Engineering and Computer Science, University of Central Florida
To appear in: Leonardo (Transactions Section) 41(1), 2007. 2 pages (This article is a short notice written for the transaction section of the journal)

Compositional Pattern Producing Networks: A Novel Abstraction of Development
Kenneth O. Stanley
School of Electrical Engineering and Computer Science, University of Central Florida
In: Genetic Programming and Evolvable Machines Special Issue on Developmental Systems New York, NY: Springer, 2007. 36 pages.

REAL-TIME NEUROEVOLUTION IN THE NERO VIDEO GAME
Kenneth O. Stanley, Bobby D. Bryant, and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
IEEE Transactions on Evolutionary Computation 9(6): 653-668, December 2005.

COMPETITIVE COEVOLUTION THROUGH EVOLUTIONARY COMPLEXIFICATION
Kenneth O. Stanley and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Journal of Artificial Intelligence Research 21: 63-100, 2004.

A TAXONOMY FOR ARTIFICIAL EMBRYOGENY
Kenneth O. Stanley and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Artificial Life journal 9(2):93-130, 2003.

EVOLVING NEURAL NETWORKS THROUGH AUGMENTING TOPOLOGIES
Kenneth O. Stanley and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Evolutionary Computation 10(2):99-127, 2002.

ONLINE INTERACTIVE NEURO-EVOLUTION
Adrian Agogino(1), Kenneth Stanley(2), and Risto Miikkulainen(2)
(1) Dept. of Electrical and Computer Engineering, The University of Texas at Austin
(2) Department of Computer Sciences, The University of Texas at Austin
Neural Processing Letters 11(1):29-37, 2000.

Conference and Symposium Papers

Exploiting Open-Endedness to Solve Problems Through the Search for Novelty
Joel Lehman and Kenneth O. Stanley
School of Electrical Engineering and Computer Science, University of Central Florida
To appear in: Proceedings of the Eleventh International Conference on Artificial Life (ALIFE XI). Cambridge MA: MIT Press, 2008. 8 pages.

Generative Encoding for Multiagent Learning
David B. D'Ambrosio and Kenneth O. Stanley
School of Electrical Engineering and Computer Science, University of Central Florida
To appear in: Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2008). New York, NY: ACM, 2007. 8 pages.
Accompanied by videos of evolved multiagent behavior.
Winner of the Best Paper Award in Generative and Developmental Systems at GECCO 2008

A Case Study on the Critical Role of Geometric Regularity in Machine Learning
Jason J. Gauci and Kenneth O. Stanley
School of Electrical Engineering and Computer Science, University of Central Florida
To appear in: Proceedings of the Twenty-Third AAAI Conference on Artificial Intelligence (AAAI-2008). Menlo Park, CA: AAAI Press, 2008 (6 pages)

Picbreeder: Evolving Pictures Collaboratively Online
Jimmy Secretan, Nicholas Beato, David B. D'Ambrosio, Adelein Rodriguez, Adam Campbell and Kenneth O. Stanley
School of Electrical Engineering and Computer Science, University of Central Florida
To appear in: Proceedings of Computer Human Interaction Conference (CHI 2008). New York, NY: ACM, 2008 (10 pages)

Scaffolding for Interactively Evolving Novel Drum Tracks for Existing Songs
Amy K. Hoover, Michael P. Rosario, and Kenneth O. Stanley
School of Electrical Engineering and Computer Science, University of Central Florida
In: Proceedings of the Sixth European Workshop on Evolutionary and Biologically Inspired Music, Sound, Art and Design (EvoMUSART 2008). New York, NY: Springer, 2008. 10 pages.
Winner of the Best Paper Award at EvoMUSART 2008

Generating Large-Scale Neural Networks Through Discovering Geometric Regularities
Jason J. Gauci and Kenneth O. Stanley
School of Electrical Engineering and Computer Science, University of Central Florida
To appear in: Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2007). New York, NY: ACM, 2007. 8 pages.

A Novel Generative Encoding for Exploiting Neural Network Sensor and Output Geometry
David B. D'Ambrosio and Kenneth O. Stanley
School of Electrical Engineering and Computer Science, University of Central Florida
To appear in: Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2007). New York, NY: ACM, 2007. 8 pages.
Nominated for Best Paper Award in the Generative and Developmental Systems track at GECCO-2007

NEAT Particles: Design, Representation, and Animation of Particle System Effects
Erin Hastings, Ratan Guha, and Kenneth O. Stanley
School of Electrical Engineering and Computer Science, University of Central Florida
To appear in: Proceedings of the IEEE Symposium on Computational Intelligence and Games (CIG'07). Piscataway, NJ: IEEE, 2007. 7 pages.

Exploiting Regularity Without Development
Kenneth O. Stanley
School of Electrical Engineering and Computer Science, University of Central Florida
To appear in: Proceedings of the AAAI Fall Symposium on Developmental Systems. Meno Park, CA: AAAI Press, 2006. 8 pages.

EVOLVING NEURAL NETWORK AGENTS IN THE NERO VIDEO GAME
Kenneth O. Stanley, Bobby D. Bryant, and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Proceedings of the IEEE 2005 Symposium on Computational Intelligence and Games (CIG'05). Piscataway, NJ: IEEE, 2005.
Winner of the Best Paper Award at CIG'05

RETAINING LEARNED BEHAVIOR DURING REAL-TIME NEUROEVOLUTION
Thomas D'Silva, Roy Janik, Michael Chrien, Kenneth O. Stanley, and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
In:Proceedings of the Artificial Intelligence and Interactive Digital Entertainment Conference (AIIDE 2005).

REAL-TIME LEARNING IN THE NERO VIDEO GAME
Kenneth O. Stanley, Ryan Cornelius, Risto Miikkulainen, Thomas D'Silva, and Aliza Gold
Department of Computer Sciences, The University of Texas at Austin
In:Proceedings of the Artificial Intelligence and Interactive Digital Entertainment Conference Demonstration Program (AIIDE 2005).
Note: This paper is a short synopsis that accompanied a live demonstration booth

NEUROEVOLUTION OF AN AUTOMOBILE CRASH WARNING SYSTEM
Kenneth O. Stanley, Nate Kohl, Rini Sherony, and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
In:Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-2005).

AUTOMATIC FEATURE SELECTION IN NEUROEVOLUTION
Shimon Whiteson, Peter Stone, Kenneth O. Stanley, Risto Miikkulainenn and Nate Kohl
Department of Computer Sciences, The University of Texas at Austin
In:Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-2005).

TOWARDS AN EMPIRICAL MEASURE OF EVOLVABILITY
Joseph Reisinger, Kenneth O. Stanley, and Risto Miikkulainenn
Department of Computer Sciences, The University of Texas at Austin
In:Proceedings of the Genetic and Evolutionary Computation Conference Workshop Program (GECCO-2005).

EVOLVING A ROVING EYE FOR GO
Kenneth O. Stanley and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-2004). New York, NY: Springer-Verlag, 2004.

EXPLOITING MORPHOLOGICAL CONVENTIONS FOR GENETIC REUSE
Kenneth O. Stanley, Joseph Reisinger, and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-2004) Workshop Program. New York, NY: Springer-Verlag, 2004
GECCO Workshop on Modularity, Regularity, and Hierarchy in Evolutionary Computation

EVOLVING REUSABLE NEURAL MODULES
Joseph Reisinger, Kenneth O. Stanley, and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-2004). New York, NY: Springer-Verlag, 2004.

AUTOMATIC FEATURE SELECTION IN NEUROEVOLUTION (older workshop version)
Shimon Whiteson, Kenneth O. Stanley, and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-2004) Workshop Program. New York, NY: Springer-Verlag, 2004
GECCO Workshop on Self-organization in Representations for Evolutionary Algorithms

EVOLVING ADAPTIVE NEURAL NETWORKS WITH AND WITHOUT ADAPTIVE SYNAPSES
Kenneth O. Stanley, Bobby D. Bryant, and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Proceedings of the 2003 IEEE Congress on Evolutionary Computation (CEC-2003). Canberra, Australia: IEEE Press, 2003.

ACHIEVING HIGH-LEVEL FUNCTIONALITY THROUGH COMPLEXIFICATION
Kenneth O. Stanley and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Proceedings of the AAAI-2003 Spring Symposium on Computational Synthesis. Stanford, CA: AAAI Press, 2003.

EFFICIENT REINFORCEMENT LEARNING THROUGH EVOLVING NEURAL NETWORK TOPOLOGIES
Kenneth O. Stanley and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-2002). San Francisco, CA: Morgan Kaufmann, 2002.
Winner of the Best Paper Award in Genetic Algorithms

CONTINUAL COEVOLUTION THROUGH COMPLEXIFICATION
Kenneth O. Stanley and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-2002). San Francisco, CA: Morgan Kaufmann, 2002.

THE DOMINANCE TOURNAMENT METHOD FOR MONITORING PROGRESS IN COEVOLUTION
Kenneth O. Stanley and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-2002) Workshop Program. San Francisco, CA: Morgan Kaufmann, 2002.
Analysis of Coevolution Workshop Link

EFFICIENT EVOLUTION OF NEURAL NETWORK TOPOLOGIES
Kenneth O. Stanley and Risto Miikkulainen
Department of Computer Sciences, The University of Texas at Austin
Proceedings of the 2002 Congress on Evolutionary Computation (CEC '02). Piscataway, NJ: IEEE, 2002.

Off-topic

The following tech report describes an algorithm for tracking concept drift that is unrelated to NEAT and my usual dissertation work:

LEARNING CONCEPT DRIFT WITH A COMMITTEE OF DECISION TREES
Kenneth O. Stanley
Department of Computer Sciences, The University of Texas at Austin
Technical Report AI-03-302, September 2003.

Contact me here:

kstanley@cs.ucf.edu .