Quantum AI and hybrid simulators for a Universal Quantum Field Computation Model

Quantum field theory (QFTh) simulators simulate physical systems using quantum circuits that process quantum information (qubits) via single field (SF) and/or quantum double field (QDF) transformation. This review presents models that classify states against pairwise particle states $|ij〉$, given their state transition (ST) probability $P_{|ij〉}$. A quantum AI (QAI) program, weighs and compares the field's distance between entangled states as qubits from their scalar field of radius $R\ge |r_{ij}|$. These states distribute across $〈R〉$ with expected probability $〈P_{\text{distribute}}〉$ and measurement outcome $〈M\left(P_{\text{distribute}}\right)〉=P_{|ij〉}$. A quantum-classical hybrid model of processors via QAI, classifies and predicts states by decoding qubits into classical bits. For example, a QDF as a quantum field computation model (QFCM) in IBM-QE, performs the doubling of $P_{|ij〉}$ for a strong state prediction outcome. QFCMs are compared to achieve a universal QFCM (UQFCM). This model is novel in making strong event predictions by simulating systems on any scale using QAI. Its expected measurement fidelity is $〈M\left(F\right)〉\ge 7/5$ in classifying states to select 7 optimal QFCMs to predict $〈M〉$'s on QFTh observables. This includes QFCMs' commonality of $〈M〉$ against QFCMs limitations in predicting system events. Common measurement results of QFCMs include their expected success probability $〈P_{\text{success}}〉$ over STs occurring in the system. Consistent results with high $F$'s, are averaged over STs as $〈P_{\text{distribute}}〉$yielding $〈P_{\text{success}}〉\ge 2/3$ performed by an SF or QDF of certain QFCMs. A combination of QFCMs with this fidelity level predicts error rates (uncertainties) in measurements, by which a $P_{|ij〉}=〈P_{\text{success}}〉\frac{<}{\sim }1$ is weighed as a QAI output to a QFCM user. The user then decides which QFCMs perform a more efficient system simulation as a reliable solution. A UQFCM is useful in predicting system states by preserving and recovering information for intelligent decision support systems in applied, physical, legal and decision sciences, including industry 4.0 systems.