aiqbench analysis
This research introduces a data-driven early-stopping mechanism for ES-HyperNEAT that significantly reduces computational waste by classifying fitness trajectories to identify stagnation. By optimizing thresholds based on initial trial data, the method retains over 90% of successful runs while cutting costs by roughly 42%, offering a more efficient alternative to general-purpose strategies like Hyperband for evolutionary algorithms. While specific numerical thresholds are domain-dependent, the core methodology of deriving stopping criteria from fitness dynamics provides a scalable framework applicable to other evolutionary search spaces prone to premature convergence.
Early-Stopping Thresholds for ES-HyperNEAT: A Data-Driven Approach from Fitness Dynamics
arXiv:2609.13533v1 Announce Type: cross Abstract: Most hyperparameter configurations for Evolvable-Substrate HyperNEAT (ES-HyperNEAT) produce networks that stagnate at random-guessing performance, wasting computational resources. We frame early stopping as binary classification on early fitness trajectories: for each trial, we compute the cumulative median of best-per-generation fitness and test it against a threshold derived by maximizing the F1 score on an initial 90-trial dataset. The resulting rule (generation G* = 3, threshold T* = 0.140) achieves F1 = 0.872 on 180 independent validation trials, retaining over 90% of successful trials while cutting computational cost by 41.6%. Compared to Hyperband, our domain-specific rule is 64% more efficient with higher mean fitness, though Hyperband occasionally discovers higher peak solutions. On a converged search population the rule becomes too aggressive (recall 31.1%), motivating adaptive thresholds. The specific thresholds are ES-HyperNEAT-specific, but the methodology, deriving stopping criteria from fitness dynamics classification, is applicable to other evolutionary algorithms with stagnation-prone hyperparameter spaces.
Abstract:Most hyperparameter configurations for Evolvable-Substrate HyperNEAT (ES-HyperNEAT) produce networks that stagnate at random-guessing performance, wasting computational resources. We frame early stopping as binary classification on early fitness trajectories: for each trial, we compute the cumulative median of best-per-generation fitness and test it against a threshold derived by maximizing the F1 score on an initial 90-trial dataset. The resulting rule (generation G* = 3, threshold T* = 0.140) achieves F1 = 0.872 on 180 independent validation trials, retaining over 90% of successful trials while cutting computational cost by 41.6%. Compared to Hyperband, our domain-specific rule is 64% more efficient with higher mean fitness, though Hyperband occasionally discovers higher peak solutions. On a converged search population the rule becomes too aggressive (recall 31.1%), motivating adaptive thresholds. The specific thresholds are ES-HyperNEAT-specific, but the methodology, deriving stopping criteria from fitness dynamics classification, is applicable to other evolutionary algorithms with stagnation-prone hyperparameter spaces.
8 pages, 5 figures, 9 tables. Accepted version of a paper published at the 2026 IEEE Congress on Evolutionary Computation (CEC), part of the IEEE World Congress on Computational Intelligence (WCCI 2026), Maastricht, Netherlands, 21-26 June 2026. (c) 2026 IEEE
Neural and Evolutionary Computing (cs.NE); Machine Learning (cs.LG)
arXiv:2609.13533 [cs.NE]
(or arXiv:2609.13533v1 [cs.NE] for this version)
https://doi.org/10.48550/arXiv.2609.13533 arXiv-issued DOI via DataCite (pending registration)
Full text under the publisher's open licence · 212 words
Related coverage
[AINews] Reality Checks on AI News (Yegge shuts down Gas Town, Databricks’ +60% Astra cost)
Latent Space · Sep 17, 2026
[AINews] High Taste Testers Ep. 1
Latent Space · Oct 17, 2026
AI Applications in Finance: A Practical Use Case Guide
Databricks · Jul 24, 2026
Reviewing failure modes in science as we prepare for AI researchers
LessWrong · Oct 8, 2026