Active Inference for Antimicrobial Resistance Mechanism Identification: Adaptive, Self-Learning Diagnostic Test Selection Using the Prometheus 1T Engine

We present the Prometheus 1T Engine, a high-performance Active Inference system for antimicrobial resistance (AMR) mechanism identification. The engine implements exact Bayesian inference to sequentially select optimal diagnostic tests via Expected Information Gain (EIG), and is mathematically cross-validated against the official pymdp library (UCL/Friston Lab).