A Fresh Look at Lamarckian Evolution and the Baldwin Effect
Abstract
Baldwinian and Lamarckian evolution have existed for a long time in evolutionary algorithms (EAs) without ever dominating the academic literature or practical applications. In this work, we use modern empirical and theoretical methods to revisit Lamarckian and Baldwinian evolution and rigorously compare them with the generic Darwinian evolution. On the empirical side, we run a comprehensive suite of experiments on graphs from six different datasets from the recent GraphBench benchmark on Maximum Independent Set and Maximum Cut problems. Our results show that Baldwinian and Lamarckian evolution consistently outperform Darwinian evolution, confirming the great potential of local search augmented evolutionary algorithms. Notably, in the great majority of cases, all EAs outperform recent deep learning baselines and approach the performance of highly specialised heuristic and exact solvers. We furthermore report a high-performing set of generalist parameters for all studied evolution types that we hope will be of use to practitioners in future. On the theoretical side, we extend the existing DeceptiveLeadingBlocks benchmark to arbitrary block length k. For all constant k, we then prove asymptotically tight runtime bounds for the (1+1) EA in the three evolution types on this benchmark. For Baldwinian evolution, these are independent of k, whereas for the other two evolution types, the runtimes steeply increase with growing value of k.
Versions
Related
- On the Upsilon invariant and the Baldwin-Vela-Vick b-invariant of fibered knots
- Simplicity Above Elegance: Another Look at the Asymptotically Correct Standardization of Snijders
- Where should we look for the deviation from quantum mechanics?
- LampAttention: Look-Ahead Mixed-Precision FlashAttention for Dedicated Accelerators
- When Fresh Information Is Not Enough: Age-of-Actuation-Optimal Scheduling for Energy-Harvesting Actuators
BibTeX
@misc{benito2026fresh,
title = {{A Fresh Look at Lamarckian Evolution and the Baldwin Effect}},
author = {Inès Benito and Johannes F. Lutzeyer and Benjamin Doerr},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2605.28703},
eprint = {2605.28703},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.28703},
}
RIS
TY - UNPB
TI - A Fresh Look at Lamarckian Evolution and the Baldwin Effect
AU - Inès Benito
AU - Johannes F. Lutzeyer
AU - Benjamin Doerr
PY - 2026
DA - 2026/09/15
PB - arXiv
DO - 10.48550/arxiv.2605.28703
UR - https://arxiv.org/abs/2605.28703
AB - Baldwinian and Lamarckian evolution have existed for a long time in evolutionary algorithms (EAs) without ever dominating the academic literature or practical applications. In this work, we use modern empirical and theoretical methods to revisit Lamarckian and Baldwinian evolution and rigorously compare them with the generic Darwinian evolution. On the empirical side, we run a comprehensive suite of experiments on graphs from six different datasets from the recent GraphBench benchmark on Maximum Independent Set and Maximum Cut problems. Our results show that Baldwinian and Lamarckian evolution consistently outperform Darwinian evolution, confirming the great potential of local search augmented evolutionary algorithms. Notably, in the great majority of cases, all EAs outperform recent deep learning baselines and approach the performance of highly specialised heuristic and exact solvers. We furthermore report a high-performing set of generalist parameters for all studied evolution types that we hope will be of use to practitioners in future. On the theoretical side, we extend the existing DeceptiveLeadingBlocks benchmark to arbitrary block length k. For all constant k, we then prove asymptotically tight runtime bounds for the (1+1) EA in the three evolution types on this benchmark. For Baldwinian evolution, these are independent of k, whereas for the other two evolution types, the runtimes steeply increase with growing value of k.
ER -