🤖 AI Summary
This work addresses the joint optimization of feature selection and classifier hyperparameters for classification tasks by proposing MMAO-Cls, a method grounded in a metabolic multi-agent optimization framework. MMAO-Cls simultaneously encodes binary feature masks and hyperparameters within a mixed search space, incorporating an energy mechanism, population budgeting, and lifecycle modeling to balance accuracy and complexity. It innovatively leverages feature information priors to dynamically adjust the feature budget and introduces a regularized validation reward based on subset compactness and overfitting gap. Evaluated on seven standard tabular datasets, MMAO-Cls achieves an average test accuracy of 0.8882—outperforming RandomSearch and GA-lite—and yields the most compact feature subsets with an average feature ratio of 0.4881, although the performance gain does not reach statistical significance.
📝 Abstract
This paper studies whether the Metabolic Multi-Agent Optimizer (MMAO) can act as a credible outer-loop optimizer for classification model selection. We propose MMAO-Cls, a mixed-space realization in which each agent jointly encodes a binary feature mask and classifier hyperparameters, while private energy, communal budget, role drift, and lifecycle turnover are mapped to the accuracy-complexity tradeoff of wrapper learning. The implementation is strengthened by deriving feature-budget adaptation from feature-information priors and by regularizing validation reward with both subset compactness and train-validation overfitting gap. We evaluate MMAO-Cls on seven standard tabular benchmarks with three seeds each and compare it against RandomSearch, GA-lite, PSO-lite, and an endogenous no-sharing ablation. On the aggregate validation objective, MMAO-Cls ranks second ($0.9433$) behind GA-lite ($0.9446$). On held-out test performance, it reaches mean score $0.8882$, improving over RandomSearch ($0.8808$) and GA-lite ($0.8857$), remaining close to PSO-lite ($0.8874$) and the no-sharing ablation ($0.8900$), while using the most compact mean held-out feature subset among all compared methods (feature ratio $0.4881$). Pairwise tests show that these margins are not yet statistically significant. The resulting claim is therefore conservative: MMAO-Cls supports classification applicability and compact mixed-space search more clearly than it isolates communal sharing as a decisive standalone advantage.