analiza aiqbench
This study rigorously demonstrates that evolutionary algorithms incorporating Baldwinian or Lamarckian mechanisms consistently outperform standard Darwinian approaches and recent deep learning baselines on complex graph optimization problems. Theoretically, the research proves that these local search-augmented methods offer superior asymptotic runtime performance compared to generic evolutionary strategies. These findings provide a robust theoretical foundation and practical parameter sets for practitioners seeking to enhance optimization capabilities in AI systems.
A Fresh Look at Lamarckian Evolution and the Baldwin Effect
arXiv:2605.28703v2 Announce Type: replace-cross 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 Deceptive Leading Block benchmark to arbitrary block length and use tools from modern theoretical runtime analysis to prove upper and lower bounds on the expected runtime. For block lengths greater than two, Baldwinian evolution is asymptotically faster than Lamarckian which is asymptotically faster than Darwinian evolution. When accounting for the cost of the local search procedure in fitness evaluations, the ordering depends on the implementation with Baldwinian evolution staying fastest from small block lengths onwards, explaining its strong empirical performance.
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 Deceptive Leading Block benchmark to arbitrary block length and use tools from modern theoretical runtime analysis to prove upper and lower bounds on the expected runtime. For block lengths greater than two, Baldwinian evolution is asymptotically faster than Lamarckian which is asymptotically faster than Darwinian evolution. When accounting for the cost of the local search procedure in fitness evaluations, the ordering depends on the implementation with Baldwinian evolution staying fastest from small block lengths onwards, explaining its strong empirical performance.
Neural and Evolutionary Computing (cs.NE); Artificial Intelligence (cs.AI); Data Structures and Algorithms (cs.DS); Optimization and Control (math.OC)
arXiv:2605.28703 [cs.NE]
(or arXiv:2605.28703v2 [cs.NE] for this version)
In Proceedings of the 19th International Conference on Parallel Problem Solving from Nature (PPSN XIX), Lecture Notes in Computer Science, vol. 16988, Springer, pp. 459-474, 2026
https://doi.org/10.1007/978-3-032-36229-2_28 DOI(s) linking to related resources
Text integral sub licența deschisă a publicației · 295 cuvinte
Subiecte conexe
Microsoft’s AI chief says superintelligence is near, but won’t take your job
The Verge · 8 iun. 2026
Presentation: Adaptive Recommenders in the Real World: Inference, Evals, and System Design
InfoQ · 26 sept. 2026
Enabling Evolutionary Database Development: database branching with Lakebase, continued
Databricks · 5 iun. 2026
Does preservation make sense before we know how to revive?
LessWrong · 15 iun. 2026