All Publications


2009

Online Learning of Assignments
with Daniel Golovin and Andreas Krause, in Proceedings of the Twenty-Third Annual Conference on Neural Information Processing Systems (NIPS 2009)
[Paper] [Longer technical report]

Tighter Bounds for Multi-Armed Bandits with Expert Advice
with Brendan McMahan, in Proceedings of the Twenty-Second Conference on Computational Learning Theory (COLT 2009)
[Paper]

2008

An Online Algorithm for Maximizing Submodular Functions
with Daniel Golovin, in Proceedings of the Twenty-Second Annual Conference on Neural Information Processing Systems (NIPS 2008)
[Paper] [Longer technical report]

New Techniques for Algorithm Portfolio Design
with Stephen Smith, in Proceedings of the Twenty-Fourth Conference on Uncertainty in Artificial Intelligence (UAI 2008)
[Paper]

2007

Combining Multiple Heuristics Online
with Daniel Golovin and Stephen Smith, in Proceedings of the Twenty-Second Conference on Artificial Intelligence (AAAI 2007)
[Paper] [Slides (PDF,QuickTime)]

Restart Schedules for Ensembles of Problem Instances
with Daniel Golovin and Stephen Smith, in Proceedings of the Twenty-Second Conference on Artificial Intelligence (AAAI 2007)
[Paper] [Slides (PDF,QuickTime)]

Using Decision Procedures Efficiently for Optimization
with Stephen Smith, in Proceedings of the Seventeenth International Conference on Automated Planning and Scheduling (ICAPS 2007)
[Paper] [Slides (PDF,QuickTime)]

2006

A Simple Distribution-Free Approach to the Max k-Armed Bandit Problem
with Stephen Smith, in Proceedings of the Twelfth International Conference on Principles and Practice of Constraint Programming (CP 2006)
[Paper] [Slides]

An Asymptotically Optimal Algorithm for the Max k-Armed Bandit Problem
with Stephen Smith, in Proceedings of the Twenty-First National Conference on Artificial Intelligence (AAAI 2006)
[Paper] [Slides]
A preliminary version appeared as a CMU technical report.

How the Landscape of Random Job Shop Scheduling Instances Depends on the Ratio of Jobs to Machines
with Stephen Smith, in Journal of Artificial Intelligence Research 26
[Paper]
A preliminary version appeared as a CMU technical report.

Exploiting the Power of Local Search in a Branch and Bound Algorithm for Job Shop Scheduling
with Stephen Smith, in Proceedings of the Sixteenth International Conference on Automated Planning and Scheduling (ICAPS 2006)
[Paper] [Slides]

2005

Characterizing the Distribution of Low-Makespan Schedules in the Job Shop Scheduling Problem
with Stephen Smith, in Proceedings of the Fifteenth International Conference on Automated Planning and Scheduling (ICAPS 2005)
[Paper] [Slides]

Improved General-Purpose Controllers
with Martin Keane and John Koza, U.S. Patent 6,847,851
[Patent]

2004

Upper Bounds on the Time and Space Complexity of Optimizing Additively Separable Functions
in Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2004)
[Paper] [Slides] [Source code, etc.]

Routine High-Return Human-Competitive Evolvable Hardware
with John Koza and Martin Keane, in Proceedings of the NASA/DoD Conference on Evolvable Hardware (EH-2004)
[Paper]

Invention and Creativity in Automated Design by Means of Genetic Programming
with John Koza, Martin Keane, Thom Adams, and Lee Jones, AI EDAM, 18(3):245-269
[Abstract]

Toward Industrial Strength Automated Design of Analog Electrical Circuits by Means of Genetic Programming
with John Koza, Lee Jones, and Martin Keane, in Genetic Programming Theory and Practice II (GPTP 2004)
[Paper]

Routine Automated Synthesis of Five Patented Analog Circuits using Genetic Programming
with John Koza and Martin Keane, Applied Soft Computing 8(5):318-324

2003

Automated Discovery of Numerical Approximation Formulae via Genetic Programming
in Genetic Programming and Evolvable Machines, 4(3):255-286
[Abstract]

Routine High-Return Human-Competitive Machine Learning
with John Koza and Martin Keane, in Proceedings of the 2003 International Conference on Machine Learning and Applications (ICMLA 2003)
[Paper]

Routine Human-Competitive Machine Intelligence by Means of Genetic Programming
with John Koza and Martin Keane, in Proceedings of SPIE 2003
[Paper]

Two Broad Classes of Functions for which a No Free Lunch Result Does Not Hold
in Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2003)
[Paper] [Slides]

The Importance of Reuse and Development in Evolvable Hardware
with John Koza and Martin Keane, in Proceedings of the NASA/DoD Conference on Evolvable Hardware (EH-2003)
[Paper]

Genetic Programming IV: Routine Human-Competitive Machine Intelligence
with John Koza, Martin Keane, Bill Mydlowec, Jessen Yu, and Guido Lanza, Klumer Academic Publishers
[Description] [Chapter 1]

What's AI Done for Me Lately? Genetic Programming's Human-Competitive Results
with John Koza and Martin Keane, in IEEE Intelligent Systems
[Paper]

Use of Genetic Programming for Automatic Synthesis of Post-2000 Patented Analog Electrical Circuits and Patentable Controllers
with Martin Keane and John Koza, in Computed Aided Optimum Design of Structures VII
[Paper] [Slides]

Automated Synthesis by Means of Genetic Programming of Complex Structures Incorporating Reuse, Hierarchies, Development, and Parameterized Topologies
with John Koza and Martin Keane, in Genetic Programming Theory and Practice (GPTP 2003)
[Paper]

The Root Causes of Code Growth in Genetic Programming
in Proceedings of EuroGP 2003
[Paper]

Automatic Synthesis using Genetic Programming of Improved PID Tuning Rules
with Martin Keane and John Koza, in Proceedings of the 2003 Intelligent Control Systems and Signal Processing Conference (ICONS 2003)
[Paper] [Slides]

Automated Synthesis by Means of Genetic Programming of Human-Competitive Designs Employing Reuse, Hierarchies, Modularities, Development, and Parameterized Topologies
with John Koza and Martin Keane, in Computational Synthesis: From Basic Building Blocks to High Level Functionality, Papers from the 2003 AAAI Spring Symposium
[Paper]

Automatic Synthesis using Genetic Programming of Both the Topology and Sizing for Five Post-2000 Patented Analog and Mixed Analog-Digital Circuits
with Martin Keane and John Koza, in Proceedings of the 2003 Southwest Symposium on Mixed-Signal Design (SSMSD 2003)
[Paper] [Slides]

Evolving Inventions
with John Koza and Martin Keane, in Scientific American
[Abstract]

2002

Iterative Refinement of Computational Circuits using Genetic Programming
with Martin Keane and John Koza, in Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2002)
[Paper] [Slides]

Automatic Synthesis using Genetic Programming of an Improved General-Purpose Controller for Industrially-Representative Plants
with Martin Keane and John Koza, in Proceedings of the NASA/DoD Conference on Evolvable Hardware
[Paper]

Routine Duplication of Post-2000 Patented Inventions by Means of Genetic Programming
with Martin Keane and John Koza, in Proceedings of EuroGP 2002
[Paper] [Slides]

2001

Automated Discovery of Numerical Approximation Formulae via Genetic Programming
with Lee Becker, in Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2001)
[Paper] [Slides]

Toward a Better Sine Wave
with Lee Becker, in GECCO 2001 Late-Breaking Papers
[Paper] [Slides]

Automated Discovery of Numerical Approximation Formulae via Genetic Programming
Master's thesis, Computer Science, Worcester Polytechnic Institute
[Thesis]

NVIS: An Interactive Visualization Tool for Neural Networks
with Matt Ward and Sergio Alvarez, in Proceedings of SPIE 2001
[Paper]