Neural network representation and optimization of thermoelectric states of multiple interacting quantum dots
We perform quantum master equation calculations and machine learning to investigate thermoelectric properties of multiple interacting quantum dots (MQD), including electrical conductance, Seebeck coefficient, thermal conductance and the figure of merit (ZT). We show that through learning from the data obtained from the QME, the thermoelectric states of MQD can be well represented by a two-layer neural network. We also show that after training, the neural network is able to predict the thermoelectric properties of MQD with much less computational cost than QME. Based on the neural network, we further optimize MQD to achieve high ZT and power factor. This work presents a powerful route to study, represent, and optimize interacting quantum many-body systems.