SGEMAS: A Self-Growing Ephemeral Multi-Agent System for Unsupervised Online Anomaly Detection

The article introduces SGEMAS, a bio-inspired architecture for physiological signal monitoring that treats intelligence as a dynamic thermodynamic process. It uses a structural plasticity mechanism and a variational free energy objective to evolve and minimize prediction error with extreme sparsity.

Details

The article presents SGEMAS (Self-Growing Ephemeral Multi-Agent System), a novel bio-inspired architecture for physiological signal monitoring that treats intelligence as a dynamic thermodynamic process. SGEMAS uses a structural plasticity mechanism, where agents can be born and die, coupled with a variational free energy objective to naturally evolve and minimize prediction error with extreme sparsity. An ablation study on the MIT-BIH Arrhythmia Database shows that adding a multi-scale instability index to the agent dynamics significantly improves performance. In a challenging inter-patient, zero-shot setting, the final SGEMAS v3.3 model achieves a mean AUC of 0.570 ± 0.070, outperforming both its simpler variants and a standard autoencoder baseline. This result validates that a physics-based, energy-constrained model can achieve robust unsupervised anomaly detection, offering a promising direction for efficient biomedical AI.