Sign in Subscribe
Genetic research

Advances in Gene Expression Programming for Complex Optimization Problems

Strategic Insights

— 1 min read

Researchers from Shenyang Aerospace University have developed a novel algorithm to address the limitations of traditional Gene Expression Programming (GEP) in handling high-dimensional and complex optimization problems. The new algorithm, Dynamic Gene Expression Programming (DGEP), utilizes dynamic genetic operators to maintain population diversity and prevent premature convergence. A study evaluating DGEP's performance against standard GEP and alternative variants demonstrated significant improvements in fitness outcomes, R-2 values, population diversity, and the avoidance of local optima.

Key Takeaways:

  • DGEP is a novel algorithm that dynamically adjusts genetic operators to maintain population diversity and prevent premature convergence.
  • The algorithm employs two unique operators: Adaptive Regeneration Operator (DGEP-R) and Dynamically Adjusted Mutation Operator (DGEP-M).
  • An extensive evaluation of DGEP against standard GEP, NMO-SARA, and MS-GEP-A demonstrated significant improvements in fitness outcomes, R-2 values, population diversity, and the avoidance of local optima.
  • DGEP achieved optimal results for 8 benchmark functions, producing 15.7% better R-2 scores and 2.3 times larger population diversity compared to standard GEP.
  • The escape rate from local optima within DGEP was found to be 35% higher than what standard GEP could achieve.
  • The study concluded that adaptive genetic methods strengthen evolutionary procedures for solving complex problems effectively.
  • The research was supported by the Shenyang Aerospace University Innovation and Entrepreneurship Plan.
  • The study was conducted by researchers Yiping Teng, Kejia Liu, and Fang Liu at Shenyang Aerospace University.

Statistics:

  • DGEP demonstrated 15.7% better R-2 scores compared to standard GEP.
  • DGEP achieved 2.3 times larger population diversity compared to standard GEP.
  • The escape rate from local optima within DGEP was 35% higher than what standard GEP could achieve.

Sources:

  • A Study of Gene Expression Programming Algorithm for Dynamically Adjusting the Parameters of Genetic Operators. PLOS One, 2025;20(6).
  • Public Library Science, 1160 Battery Street, Ste 100, San Francisco, CA 94111, USA.
  • (Public Library of Science - www.plos.org; PLOS One - www.plosone.org)
  • Shenyang Aerospace University, School of Computer Science, Shenyang, People's Republic of China.

Read more