DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Avoiding excess computation in asynchronous evolutionary algorithms

Journal Article · · Expert Systems
DOI: https://doi.org/10.1111/exsy.13100 · OSTI ID:1880698
 [1];  [2];  [3];  [4];  [2];  [5];  [2];  [6]
  1. MITRE Corporation McLean Virginia USA, Department of Computer Science George Mason University Virginia USA
  2. Computational and Computer Science Directorate Oak Ridge National Laboratory Oak Ridge Tennessee USA
  3. Department of Electrical Engineering and Computer Science University of Tennessee Knoxville Tennessee USA
  4. Algorithms, Combinatorics, and Optimization, Pacific Northwest National Laboratory Richland Washington USA
  5. Department of Electrical and Computer Engineering George Mason University Virginia USA
  6. Department of Computer Science George Mason University Virginia USA

Abstract Asynchronous evolutionary algorithms are becoming increasingly popular as a means of making full use of many processors while solving computationally expensive search and optimization problems. These algorithms excel at keeping large clusters fully utilized, but may sometimes inefficiently sample an excess of fast‐evaluating solutions at the expense of higher‐quality, slow‐evaluating ones. We have previously introduced a steady‐state parent selection strategy, SWEET (“Selection whilE EvaluaTing”), that sometimes selects individuals that are still being evaluated and allows them to reproduce early. We perform a takeover‐time analysis that confirms that this strategy gives slow‐evaluating individuals that have higher fitnesses an increased ability to multiply in the population. We also find that SWEET appears effective at improving optimization performance on problems in which solution quality is positively correlated with evaluation time. We evaluate our approach on six simulated real‐valued optimization problems and three real‐world applications: an autonomous vehicle controller problem that involves tuning a spiking neural network and two adversarial EA problems. We further evaluate SWEET versus a basic asynchronous process in a simulated setting. We present evidence that SWEET outperforms basic asynchronous processes in a use‐case in which performance is positively correlated with evaluation time, and performs comparably (and often better) than basic asynchronous processes in several use‐cases where performance is negatively correlated with evaluation time. That said, in the cases where performance and evaluation time are negatively correlated the variance of outcomes for SWEET is notably high.

Research Organization:
Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States); Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)
Sponsoring Organization:
US Air Force Office of Scientific Research (AFOSR); USDOE; USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
Grant/Contract Number:
AC05-00OR22725; AC05-76RL01830
OSTI ID:
1880698
Report Number(s):
PNNL-SA--171021; e13100
Journal Information:
Expert Systems, Journal Name: Expert Systems Journal Issue: 5 Vol. 40; ISSN 0266-4720
Publisher:
Wiley-BlackwellCopyright Statement
Country of Publication:
United Kingdom
Language:
English

References (37)

Median-Selection for Parallel Steady-State Evolution Strategies book January 2000
Analysis of Evolutionary Algorithms in Dynamic and Stochastic Environments book November 2019
A Study on Efficient Asynchronous Parallel Multi-objective Evolutionary Algorithm with Waiting Time Limitation book January 2020
On the Performance of Master-Slave Parallelization Methods for Multi-Objective Evolutionary Algorithms book January 2013
Q-learning journal May 1992
Fixed-Target Runtime Analysis journal November 2021
Unified reinforcement Q-learning for mean field game and control problems journal January 2022
Analysis of semi-asynchronous multi-objective evolutionary algorithm with different asynchronies journal May 2019
Fuzzy Q-Learning for multi-agent decentralized energy management in microgrids journal June 2018
Fish growth trajectory tracking using Q-learning in precision aquaculture journal March 2022
A survey of federated learning for edge computing: Research problems and solutions journal June 2021
Evolutionary vs imitation learning for neuromorphic control at the edge* journal January 2022
Time and Individual Duration in Genetic Programming journal January 2020
Simulation-Based Evolutionary Optimization of Air Traffic Management journal January 2020
Asynchronous Online Federated Learning for Edge Devices with Non-IID Data conference December 2020
Effect of global parallelism on the behavior of a steady state genetic algorithm for design optimization
  • Rasheed, K.; Davison, B. D.
  • 1999. Congress on Evolutionary Computation-CEC99, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406) https://doi.org/10.1109/CEC.1999.781979
conference January 1999
Using a Queue Genetic Algorithm to Evolve Xpilot Control Strategies on a Distributed System conference January 2006
Asynchronous Evolutionary Multi-Objective Algorithms with heterogeneous evaluation costs
  • Yagoubi, Mouadh; Thobois, Ludovic; Schoenauer, Marc
  • 2011 IEEE Congress on Evolutionary Computation (CEC), 2011 IEEE Congress of Evolutionary Computation (CEC) https://doi.org/10.1109/CEC.2011.5949593
conference June 2011
Tool sequence optimization using synchronous and asynchronous parallel multi-objective evolutionary algorithms with heterogeneous evaluations conference June 2013
PasMoQAP: A parallel asynchronous memetic algorithm for solving the Multi-Objective Quadratic Assignment Problem conference June 2017
Mathematical Model of Asynchronous Parallel Evolutionary Algorithm to Analyze Influence of Evaluation Time Bias conference December 2019
Comparison of reinforcement learning algorithms applied to the cart-pole problem conference September 2017
A study of master-slave approaches to parallelize NSGA-II
  • Durillo, Juan J.; Nebro, Antonio J.; Luna, Francisco
  • Distributed Processing Symposium (IPDPS), 2008 IEEE International Symposium on Parallel and Distributed Processing https://doi.org/10.1109/IPDPS.2008.4536375
conference April 2008
Search Progress Dependent Parent Selection for Avoiding Evaluation Time Bias in Asynchronous Parallel Multi-Objective Evolutionary Algorithms conference December 2020
Neuronlike adaptive elements that can solve difficult learning control problems journal September 1983
Understanding Simple Asynchronous Evolutionary Algorithms
  • Scott, Eric O.; De Jong, Kenneth A.
  • FOGA '15: Foundations of Genetic Algorithms XIII, Proceedings of the 2015 ACM Conference on Foundations of Genetic Algorithms XIII https://doi.org/10.1145/2725494.2725509
conference January 2015
Evaluation-Time Bias in Quasi-Generational and Steady-State Asynchronous Evolutionary Algorithms
  • Scott, Eric O.; De Jong, Kenneth A.
  • GECCO '16: Genetic and Evolutionary Computation Conference, Proceedings of the Genetic and Evolutionary Computation Conference 2016 https://doi.org/10.1145/2908812.2908934
conference July 2016
Parallel evolutionary algorithm with interleaving generations
  • Pilát, Martin; Neruda, Roman
  • GECCO '17: Genetic and Evolutionary Computation Conference, Proceedings of the Genetic and Evolutionary Computation Conference https://doi.org/10.1145/3071178.3071309
conference July 2017
Library for evolutionary algorithms in Python (LEAP)
  • Coletti, Mark A.; Scott, Eric O.; Bassett, Jeffrey K.
  • GECCO '20: Genetic and Evolutionary Computation Conference, Proceedings of the 2020 Genetic and Evolutionary Computation Conference Companion https://doi.org/10.1145/3377929.3398147
conference July 2020
A Small, Low Cost Event-Driven Architecture for Spiking Neural Networks on FPGAs conference July 2020
Diagnosing autonomous vehicle driving criteria with an adversarial evolutionary algorithm
  • Coletti, Mark A.; Gao, Shang; Paulissen, Spencer
  • GECCO '21: Genetic and Evolutionary Computation Conference, Proceedings of the Genetic and Evolutionary Computation Conference Companion https://doi.org/10.1145/3449726.3459573
conference July 2021
A Survey on Recent Progress in the Theory of Evolutionary Algorithms for Discrete Optimization journal December 2021
Neuromorphic Computing for Autonomous Racing conference October 2021
Troubleshooting deep-learner training data problems using an evolutionary algorithm on Summit journal May 2020
Asynchronous Master-Slave Parallelization of Differential Evolution for Multi-Objective Optimization journal May 2013
Evolutionary model discovery of causal factors behind the socio-agricultural behavior of the Ancestral Pueblo journal December 2020
Interleaving Generation Evolutionary Algorithm with Precedence Evaluation of Tentative Offspring conference September 2020