An epigenetic-inspired evolutionary framework for dynamic optimization landscapes

Aruna Bharathi Mathangi, Mani Sai Aashrith Atkuri, Sai Praneet Yamijala, Balajapalli Venkata Satya Sriram Kumar, Abhishek Kulkarni, Raghuvardhan Kanti

Abstract


Evolutionary algorithms have long drawn from Darwinian ideas, yet they tend to struggle when the fitness landscape keeps shifting beneath them. This paper introduces the epigenetic-inspired evolutionary framework (EIEF), a new approach that borrows from the way living cells regulate gene expression to make evolution more responsive and resilient. Rather than relying solely on mutation and crossover, EIEF separates what an individual “knows “genetically from how it currently “behaves” phenotypically, using an epigenetic regulation layer that can adapt quickly without permanently altering the underlying genome. We develop three core mechanisms: dynamic methylation masking (DMM), which adjusts how strongly each gene influences behaviour based on environmental feedback; chromatin remodelling crossover (CRC), a recombination operator that respects regulatory context; and transgenerational epigenetic inheritance (TEI), which passes acquired regulatory knowledge across generations. We evaluate EIEF against six competing algorithms across 14 dynamic benchmark problems spanning continuous optimization, protein folding, logistics scheduling, and adaptive control. The results show that EIEF tracks shifting optima more precisely, recovers diversity faster after landscape changes, and maintains this advantage as problem complexity scales up. We also provide formal guarantees on diversity preservation and tracking speed, grounding the empirical improvements in theory.

Keywords


Chromatin remodeling; Dynamic optimization; Epigenetics; Evolutionary algorithms; Fitness landscape tracking; Gene regulation; Transgenerational inheritance

Full Text:

PDF


DOI: https://doi.org/10.11591/ehs.v4i2.pp146-154

Refbacks

  • There are currently no refbacks.


Copyright (c) 2026 Aruna Bharathi Mathangi, Mani Sai Aashrith Atkuri, Sai Praneet Yamijala, Balajapalli Venkata Satya Sriram Kumar, Abhishek Kulkarni, Raghuvardhan Kanti

CC BY-SA 4.0

Intellectual Journal of Energy Harvesting and Storage is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

View This Journal Visitor Statistics