Gregory R. Kramer

Wright State University, Dayton, Ohio, United States

Are you Gregory R. Kramer?

Claim your profile

Publications (6)3.65 Total impact

  • G.R. Kramer · J.C. Gallagher
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, the authors present a performance analysis of a mini-population evolutionary algorithm (EA) on a robot-locomotion control problem. The results indicate that the nature of the search space allows for the design of highly efficient search algorithms that could greatly outperform current hardware-amenable techniques. The authors additionally believe that these search characteristics may be inherent in many practical problems, making the results useful for the larger community.
    Evolutionary Computation, 2005. The 2005 IEEE Congress on; 10/2005
  • G.R. Kramer · J.C. Gallagher · M. Raymer
    [Show abstract] [Hide abstract]
    ABSTRACT: Previous work has demonstrated the efficacy and feasibility of the simulated population *cGA family for embedded EH applications. This paper introduces and discusses a new *cGA variant that increases search efficacy on a specific class of EH problems without increasing the amount of hardware required for implementation. We discuss the new EA variant and provide experimental verification of its superiority against both the Dejong benchmarks and a practical problem in the construction of an EH controller to suppress thermoacoustic instability. We also compare these results with similar population-based and a population-less EAs to help understand the effects of introducing a simulated population. We conclude with a brief discussion of possible implications for the EH community in general.
    Evolvable Hardware, 2004. Proceedings. 2004 NASA/DoD Conference on; 07/2004
  • John C. Gallagher · Saranyan Vigraham · Gregory R. Kramer
    [Show abstract] [Hide abstract]
    ABSTRACT: For many evolvable hardware applications, small size and power efficiency are critical design considerations. One manner in which significant memory, and thus, power and space savings can be realized in a hardware-based evolutionary algorithm is to represent populations of candidate solutions as probability vectors rather than as sets of bit strings. The compact genetic algorithm (CGA) is a probability vector-based evolutionary algorithm that can be efficiently and elegantly implemented in digital hardware. Unfortunately, the CGA is a very weak, first order, evolutionary algorithm that is unlikely to possess sufficient search power to enable intrinsic evolvable hardware applications. In this paper, we further develop a number of modifications to the basic CGA that significantly improve its search efficacy without substantially increasing the size and complexity of its hardware implementation. The paper provides both benchmark results demonstrating increased efficacy and a conceptual data path/microcontroller design suitable for implementation in digital hardware. Following, it demonstrates efficient implementation by making a head-to-head comparison of field programmable gate array implementations of both the classic CGA and a member of our family of modifications. The paper concludes with a discussion of future research, including several additional extensions that we expect will further increase search efficacy without increasing implementation cost.
    IEEE Transactions on Evolutionary Computation 04/2004; 8(2):111-126. DOI:10.1109/TEVC.2003.820662 · 3.65 Impact Factor
  • G.R. Kramer · J.C. Gallagher
    [Show abstract] [Hide abstract]
    ABSTRACT: Recently, we proposed a neuromorphic intrinsic online evolvable hardware (EH) system designed to learn control laws of physical devices. Since we intend to eventually build this device using mixed signal VLSI techniques, and because we intend to address control applications in which small size and low power consumption are critical, we are extremely concerned with the design of physically compact devices. This paper focuses on the evolutionary algorithm (EA) portion of our proposed system. We discuss modifications to our previously reported *CGA that significantly increases its performance against dynamic optimization problems without significantly increasing the amount of hardware required for implementation. We demonstrate the efficacy of our improvement by testing against two series of moving peak benchmarks. We conclude with discussions of both the implications of our findings and our plans for future work.
    Evolvable Hardware, 2003. Proceedings. NASA/DoD Conference on; 08/2003
  • Gregory R. Kramer · John C. Gallagher
    [Show abstract] [Hide abstract]
    ABSTRACT: Fault-tolerance, complex structure management andreconfiguration are seen as valuable characteristics.Embryonic arrays represent one novel approach that takesinspiration from nature to improve upon standardtechniques. An existing BAE SYSTEMS RASCALý ...
    5th NASA / DoD Workshop on Evolvable Hardware (EH 2003), 9-11 July 2002, Chicago, IL, USA; 01/2003
  • Source
    Gregory R. Kramer · John C. Gallagher
    [Show abstract] [Hide abstract]
    ABSTRACT: Our results show that for the single-leg locomotion problem, hypermutation increases the quality of the mrCGA’s solution in a dynamic environment, whereas the random immigrant variant produces slightly lower scores. Both of these variants can be easily added to the existing mrCGA hardware implementation without significantly increasing its complexity. In the future we plan to categorize the effects of the hypermutation and random immigrant strategies on the mrCGA for a variety of generalized benchmarks. This categorization will be useful to help determine which dynamic optimization strategy should be employed for a given problem.
    Genetic and Evolutionary Computation - GECCO 2003, Genetic and Evolutionary Computation Conference, Chicago, IL, USA, July 12-16, 2003. Proceedings, Part I; 01/2003