Non Equilibrium Green Function

Non-equilibrium Green's function (NEGF) methods are a powerful computational technique for simulating quantum transport in nanoscale devices, aiming to accurately predict their electronic and transport properties. Current research focuses on mitigating the high computational cost of NEGF simulations through the integration of machine learning, such as convolutional generative networks and end-to-end differentiable models implemented in frameworks like PyTorch, to accelerate convergence and enable efficient sensitivity analysis and inverse design. These advancements significantly improve the feasibility of NEGF simulations for high-throughput applications in materials science and nanoelectronics design, facilitating faster development cycles and optimization of novel devices.

Papers