Fig. 1: Overview of AI creating nanopores for efficient water desalination via the integration of CNN and DRL.

The whole framework runs by removing atoms sequentially. At each timestep t, at most one candidate atom (colored as red) is removed from the current graphene nanopore g_t to generate a updated nanopore g_t+1. Any dangling atoms caused by the removal of candidate atom are also removed from g_t. g_t+1 is fed into a CNN-based performance predictor to predict water flux f_t+1 and ion rejection rate i_t+1. Meanwhile, the geometrical feature is extracted from the CNN. The reward is then calculated from the predicted i_t+1 and f_t+1. The geometrical feature is concatenated with the fingerprint and atom coordinates as the state s_t+1. Given g_t+1, candidate atoms to remove are picked from those located at the edge of the nanopore. The DRL agent constructed upon deep Q-network takes the reward, candidate atoms, and state as input to determine the next atom to remove from the graphene.