R. Paul Wiegand
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
R. Paul Wiegand has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Alessio Gaspar
University of South Florida, Tampa, USA
2 shared publications - Dmytro Vitel
University of South Florida
2 shared publications - Annie S. Wu
University of Central Florida
1 shared publication - Anthony Bucci
Brandeis University, Waltham, MA
1 shared publication - Brian Goldiez
Universidad Central de la Florida
1 shared publication - Chandana Ghosh
Institute for Simulation & Training, Orlando, FL
1 shared publication
Latest Publications
- Open Access
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1081-1089
Coordinate System Extraction as the Search Driver in Test-Based Genetic Programming
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1081-1089
Open AccessCoordinate System Extraction as the Search Driver in Test-Based Genetic Programming
- 17 April 2025
- Published by Springer Nature
- p. 541-556
Robust Search for the Underlying Objectives in Black-Box Games with Binary Outcomes
- 17 April 2025
- Published by Springer Nature
- p. 541-556
Robust Search for the Underlying Objectives in Black-Box Games with Binary Outcomes
- 3 September 2024
- Published by MIT Press in Evolutionary Computation
- Vol. 32 (3) , 249-273
Preliminary Analysis of Simple Novelty Search
Novelty search is a powerful tool for finding diverse sets of objects in complicated spaces. Recent experiments on simplified versions of novelty search introduce the idea that novelty search happens at the level of the archive space, rather than individual points. The sparseness measure and archive update criterion create a process that is driven by a two measures: (1) spread out to cover the space while trying to remain as efficiently packed as possible, and (2) metrics inspired by k nearest neighbor theory. In this paper, we generalize previous simplifications of novelty search to include traditional population (μ,λ) dynamics for generating new search points, where the population and the archive are updated separately. We provide some theoretical guidance regarding balancing mutation and sparseness criteria and introduce the concept of saturation as a way of talking about fully covered spaces. We show empirically that claims that novelty search is inherently objectiveless are incorrect. We leverage the understanding of novelty search as an optimizer of archive coverage, suggest several ways to improve the search, and demonstrate one simple improvement—generating some new points directly from the archive rather than the parent population.
- 3 September 2024
- Published by MIT Press in Evolutionary Computation
- Vol. 32 (3) , 249-273
Preliminary Analysis of Simple Novelty Search
Novelty search is a powerful tool for finding diverse sets of objects in complicated spaces. Recent experiments on simplified versions of novelty search introduce the idea that novelty search happens at the level of the archive space, rather than individual points. The sparseness measure and archive update criterion create a process that is driven by a two measures: (1) spread out to cover the space while trying to remain as efficiently packed as possible, and (2) metrics inspired by k nearest neighbor theory. In this paper, we generalize previous simplifications of novelty search to include traditional population (μ,λ) dynamics for generating new search points, where the population and the archive are updated separately. We provide some theoretical guidance regarding balancing mutation and sparseness criteria and introduce the concept of saturation as a way of talking about fully covered spaces. We show empirically that claims that novelty search is inherently objectiveless are incorrect. We leverage the understanding of novelty search as an optimizer of archive coverage, suggest several ways to improve the search, and demonstrate one simple improvement—generating some new points directly from the archive rather than the parent population.
- Open Access
- 31 January 2023
- Published by MDPI AG in Electronics
- Vol. 12 (3) , 706
Mitigating Catastrophic Forgetting with Complementary Layered Learning
Catastrophic forgetting is a stability–plasticity imbalance that causes a machine learner to lose previously gained knowledge that is critical for performing a task. The imbalance occurs in transfer learning, negatively affecting the learner’s performance, particularly in neural networks and layered learning. This work proposes a complementary learning technique that introduces long- and short-term memory to layered learning to reduce the negative effects of catastrophic forgetting. In particular, this work proposes the dual memory system in the non-neural network approaches of evolutionary computation and Q-learning instances of layered learning because these techniques are used to develop decision-making capabilities for physical robots. Experiments evaluate the new learning augmentation in a multi-agent system simulation, where autonomous unmanned aerial vehicles learn to collaborate and maneuver to survey an area effectively. Through these direct-policy and value-based learning experiments, the proposed complementary layered learning is demonstrated to significantly improve task performance over standard layered learning, successfully balancing stability and plasticity.
- 31 January 2023
- Published by MDPI AG in Electronics
- Vol. 12 (3) , 706
Open AccessMitigating Catastrophic Forgetting with Complementary Layered Learning
Catastrophic forgetting is a stability–plasticity imbalance that causes a machine learner to lose previously gained knowledge that is critical for performing a task. The imbalance occurs in transfer learning, negatively affecting the learner’s performance, particularly in neural networks and layered learning. This work proposes a complementary learning technique that introduces long- and short-term memory to layered learning to reduce the negative effects of catastrophic forgetting. In particular, this work proposes the dual memory system in the non-neural network approaches of evolutionary computation and Q-learning instances of layered learning because these techniques are used to develop decision-making capabilities for physical robots. Experiments evaluate the new learning augmentation in a multi-agent system simulation, where autonomous unmanned aerial vehicles learn to collaborate and maneuver to survey an area effectively. Through these direct-policy and value-based learning experiments, the proposed complementary layered learning is demonstrated to significantly improve task performance over standard layered learning, successfully balancing stability and plasticity.
- 6 June 2020
- Published by Springer Nature in Swarm Intelligence
- Vol. 14 (3) , 233-258
Response probability enhances robustness in decentralized threshold-based robotic swarms
- 6 June 2020
- Published by Springer Nature in Swarm Intelligence
- Vol. 14 (3) , 233-258
Response probability enhances robustness in decentralized threshold-based robotic swarms
- 7 June 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Reliability
- Vol. 68 (4) , 1214-1226
Analytical and Evolutionary Methods for Finding Cut Volumes in Fault Trees Constrained by Location
Fault tree analysis (FTA) is used to find and mitigate vulnerabilities in a system based on its constituent components. Methods exist to efficiently find minimal cut sets (MCSs), which are combinations of components whose failure causes the system to fail. However, traditional FTA ignores the physical location of the components. Components that are close to each other could be defeated by a single event with a radius of effect, such as an explosion or fire. This motivates the search for techniques to identify such vulnerabilities. Adding physical locations to the fault tree structure can help identify vulnerabilities in the overall system. Using this information requires extending existing solution methods or developing entirely new methods. In this paper, two solution approaches were explored. The first executes traditional FTA software, then searches for clusters in the resulting MCS to find these vulnerabilities. The second uses an evolutionary algorithm to search directly for volumes containing components that form cut sets. Results show that the evolutionary approach provided better answers (i.e., smaller volumes) overall and is suitable to identify vulnerabilities caused by proximity of components. However, the cluster approach performed well when evaluating higher numbers of locations and may be suitable in specific situations. Potential refinements to both methods are discussed.
- 7 June 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Reliability
- Vol. 68 (4) , 1214-1226
Analytical and Evolutionary Methods for Finding Cut Volumes in Fault Trees Constrained by Location
Fault tree analysis (FTA) is used to find and mitigate vulnerabilities in a system based on its constituent components. Methods exist to efficiently find minimal cut sets (MCSs), which are combinations of components whose failure causes the system to fail. However, traditional FTA ignores the physical location of the components. Components that are close to each other could be defeated by a single event with a radius of effect, such as an explosion or fire. This motivates the search for techniques to identify such vulnerabilities. Adding physical locations to the fault tree structure can help identify vulnerabilities in the overall system. Using this information requires extending existing solution methods or developing entirely new methods. In this paper, two solution approaches were explored. The first executes traditional FTA software, then searches for clusters in the resulting MCS to find these vulnerabilities. The second uses an evolutionary algorithm to search directly for volumes containing components that form cut sets. Results show that the evolutionary approach provided better answers (i.e., smaller volumes) overall and is suitable to identify vulnerabilities caused by proximity of components. However, the cluster approach performed well when evaluating higher numbers of locations and may be suitable in specific situations. Potential refinements to both methods are discussed.
- 1 January 2012
- Published by European Alliance for Innovation n.o.
An architecture supporting large scale MMOGs
The growing popularity of large-scale, highly interactive virtual reality systems such as massively multiplayer online games (MMOGs) necessitates highly robust and efficient architectures. Distributed implementations are common, but they must deal with challenges such as supporting very large numbers of closely interacting users, the need to maintain robustness in the face of hardware failure, balancing the processing load, reducing user latency, and minimizing thrashing effects caused by movement between servers. Although a number of existing techniques address each of these independently, there are no unified methods that attack these problems cohesively. We present methods to simultaneously address these critical challenges---a novel approach and associated software design intended for distributed high performance computing facilities in which the world is divided into a regular lattice of overlapping cells (providing redundancy), which are dynamically assigned to servers within the High Performance Computing (HPC) (facilitating load balancing). We believe this architecture can be applied to non-spatial cells. This architecture is currently being implemented in a test bed for further experimentation.
- 1 January 2012
- Published by European Alliance for Innovation n.o.
An architecture supporting large scale MMOGs
The growing popularity of large-scale, highly interactive virtual reality systems such as massively multiplayer online games (MMOGs) necessitates highly robust and efficient architectures. Distributed implementations are common, but they must deal with challenges such as supporting very large numbers of closely interacting users, the need to maintain robustness in the face of hardware failure, balancing the processing load, reducing user latency, and minimizing thrashing effects caused by movement between servers. Although a number of existing techniques address each of these independently, there are no unified methods that attack these problems cohesively. We present methods to simultaneously address these critical challenges---a novel approach and associated software design intended for distributed high performance computing facilities in which the world is divided into a regular lattice of overlapping cells (providing redundancy), which are dynamically assigned to servers within the High Performance Computing (HPC) (facilitating load balancing). We believe this architecture can be applied to non-spatial cells. This architecture is currently being implemented in a test bed for further experimentation.
- 1 January 2011
- Published by Springer Nature
- p. 505-527
Physicomimetic Swarm Design Considerations: Modularity, Scalability, Heterogeneity, and the Prediction Versus Control Dilemma
- 1 January 2011
- Published by Springer Nature
- p. 505-527
Physicomimetic Swarm Design Considerations: Modularity, Scalability, Heterogeneity, and the Prediction Versus Control Dilemma
- Open Access
- 7 July 2007
- Published by Association for Computing Machinery (ACM)
- p. 3133-3157
Introductory tutorial on coevolution
This tutorial is designed to introduce coevolution to those with a working familiarity with evolutionary computation. The tutorial begins by providing some basic background into what coevolution is and how it has been historically employed. The fundamental mechanics of varying types of coevolutionary algorithms are covered next, focusing heavily on issues that distinguish coevolution from traditional EC. Common pathologies are categorized and discussed, then remedies for those challenges and progress monitoring methods are introduced to the audience. The tutorial concludes with a summary and a look forward to the topics discussed in the advanced tutorial on coevolution.
- 7 July 2007
- Published by Association for Computing Machinery (ACM)
- p. 3133-3157
Open AccessIntroductory tutorial on coevolution
This tutorial is designed to introduce coevolution to those with a working familiarity with evolutionary computation. The tutorial begins by providing some basic background into what coevolution is and how it has been historically employed. The fundamental mechanics of varying types of coevolutionary algorithms are covered next, focusing heavily on issues that distinguish coevolution from traditional EC. Common pathologies are categorized and discussed, then remedies for those challenges and progress monitoring methods are introduced to the audience. The tutorial concludes with a summary and a look forward to the topics discussed in the advanced tutorial on coevolution.