Papers
8
Total Citations
108
H-Index
5
About
Moshe Sipper is a pioneer at the intersection of evolutionary computation and robotics, whose work has fundamentally reshaped how we think about machines that can learn, adapt, and even coevolve with their environments. His research spans genetic programming, adaptive robotics, and coevolutionary algorithms, with a particular focus on evolving intelligent controllers for autonomous agents. Sipper’s most influential work includes "GP-Robocode" (53 citations), where he demonstrated how genetic programming could evolve competitive virtual robot players, and "Surprise versus unsurprise" (27 citations), which explored the profound implications of emergent behaviors in robotics. He introduced the provocative concept of "adaptive environmentics"—the inverse of adaptive robotics—in which environments adapt to robots rather than the other way around. More recently, Sipper developed SAFE (Solution and Fitness Evolution), a novel coevolutionary algorithm that simultaneously evolves solutions and their own objective functions, with applications to multiobjective problems. His work on evolving sumobots and racing car controllers further showcases his talent for turning complex evolutionary concepts into tangible, competitive systems. With a career marked by both theoretical depth and playful, hands-on experimentation, Sipper continues to inspire researchers to push the boundaries of what evolution can create.
Research Focus
Key Achievements
Top Papers
- 1GP-Robocode: Using Genetic Programming to Evolve Robocode Players53 citations · 2005
- 2Surprise versus unsurprise: Implications of emergence in robotics27 citations · 2001
- 3
- 4On the origin of environments by means of natural selection6 citations · 2001
- 5GP-Sumo: Using genetic programming to evolve sumobots6 citations · 2006
- 6GP-RARS: evolving controllers for the Robot Auto Racing Simulator4 citations · 2011
- 7Romero's Pilgrimage to Santa Fe: A Tale of Robot Evolution3 citations · 1999
- 8