# Differential Evolution with Reversible Linear Transformations

arXiv: https://arxiv.org/abs/2002.02869 · HTML: https://mlsanity.com/paper/2002.02869/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Feb 2020.

## Built on by (most cited first, 8 of 8)

- [The Effects of Learning in Morphologically Evolving Robot Systems](https://mlsanity.com/paper/2111.09851/) (2021)
- [Gait-learning with morphologically evolving robots generated by L-system](https://mlsanity.com/paper/2107.08249/) (2021)
- [A comparison of controller architectures and learning mechanisms for arbitrary robot morphologies](https://mlsanity.com/paper/2309.13908/) (2023)
- [Lamarckian Inheritance Improves Robot Evolution in Dynamic Environments](https://mlsanity.com/paper/2403.19545/) (2024)
- [The importance of being constrained: dealing with infeasible solutions in Differential Evolution and beyond](https://mlsanity.com/paper/2203.03512/) (2022)
- [Lamarck's Revenge: Inheritance of Learned Traits Can Make Robot Evolution Better](https://mlsanity.com/paper/2309.13099/) (2023)
- [Exploring Robot Morphology Spaces through Breadth-First Search and Random Query](https://mlsanity.com/paper/2309.14387/) (2023)
- [Linear Proposal Operators and Stochastic Search Geometry in SOMA and Differential Evolution](https://mlsanity.com/paper/2607.29228/) (2026)

## Abstract (arXiv)

Differential evolution (DE) is a well-known type of evolutionary algorithms (EA). Similarly to other EA variants it can suffer from small populations and loose diversity too quickly. This paper presents a new approach to mitigate this issue: We propose to generate new candidate solutions by utilizing reversible linear transformation applied to a triplet of solutions from the population. In other words, the population is enlarged by using newly generated individuals without evaluating their fitness. We assess our methods on three problems: (i) benchmark function optimization, (ii) discovering parameter values of the gene repressilator system, (iii) learning neural networks. The empirical results indicate that the proposed approach outperforms vanilla DE and a version of DE with applying differential mutation three times on all testbeds.