Physics-Informed Neural Networks and Beyond: Enforcing Physical Constraints in Quantum Dissipative Dynamics

• DOI: 10.48550/arxiv.2404.14021
• Publication Date: 2024-04-22
• Authors: Arif Ullah; Yu Huang; Ming Yang; Pavlo O. Dral
• Keywords: Dynamics

Neural networks (NNs) accelerate simulations of quantum dissipative dynamics. Ensuring that these adhere to fundamental physical laws is crucial, but has been largely ignored in the state-of-the-art NN approaches. We show this may lead implausible results measured by violation trace conservation. To recover correct behavior, we develop physics-informed NNs mitigate violations a good extend. Beyond that, introduce an approach enforcing perfect conservation design.