# MMAO-Cls: Metabolic Multi-Agent Optimization for Joint Feature Selection and Classifier Tuning

arXiv: https://arxiv.org/abs/2607.01539 · HTML: https://mlsanity.com/paper/2607.01539/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 27 references that are in the map.
- First posted on arXiv 1 Jul 2026.

## Builds on (most cited first, 8 of 8)

- [ConfigX: Modular Configuration for Evolutionary Algorithms via Multitask Reinforcement Learning](https://mlsanity.com/paper/2412.07507/) (2024)
- [Large-scale Benchmarking of Metaphor-based Optimization Heuristics](https://mlsanity.com/paper/2402.09800/) (2024)
- [Optimizing with Low Budgets: a Comparison on the Black-box Optimization Benchmarking Suite and OpenAI Gym](https://mlsanity.com/paper/2310.00077/) (2023)
- [Exploratory Landscape Analysis for Mixed-Variable Problems](https://mlsanity.com/paper/2402.16467/) (2024)
- [Already Moderate Population Sizes Provably Yield Strong Robustness to Noise](https://mlsanity.com/paper/2404.02090/) (2024)
- [Comparing Optimization Algorithms Through the Lens of Search Behavior Analysis](https://mlsanity.com/paper/2507.01668/) (2025)
- [Speeding Up the NSGA-II via Dynamic Population Sizes](https://mlsanity.com/paper/2509.01739/) (2025)
- [MMAO: A Metabolic Multi-Agent Optimizer with Endogenous Resource Allocation for Continuous and Discrete Optimization](https://mlsanity.com/paper/2606.28109/) (2026)

## Abstract (arXiv)

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.