Fig. 6
From: Deep-learning-based inverse design model for intelligent discovery of organic molecules
Examples of inverse-designed host materials and their T1 values (ML: predicted value from DNN model; DFT: simulated value from DFT calculation). a Asymmetric molecules having given fragments in the training library. b Symmetric molecules where at least one fragment is different from those in the training library. c Asymmetric molecules where at least one fragment is different from those in the training library
