Table 3 The obtained results employing different configurations of the parameters that affect the exploratory behaviour of the model
From: Diversity oriented Deep Reinforcement Learning for targeted molecule generation
Results | |||||||
|---|---|---|---|---|---|---|---|
\(\tau\) | 1 | 2 | 2 | 3 | 3 | 4 | 4 |
\(\kappa\) | - | 0.70 | 0.75 | 0.70 | 0.75 | 0.70 | 0.75 |
% Desirable | 99.8 | 91.1 | 95.1 | 98.1 | 92.9 | 96.3 | 95.8 |
% Unique | 10.3 | 72.3 | 96.4 | 53.5 | 79.2 | 83.3 | 89.2 |
% Validity | 97.4 | 88.7 | 90.4 | 82.8 | 77.1 | 74.3 | 82.2 |
Diversity | 0.768 | 0.827 | 0.879 | 0.851 | 0.857 | 0.887 | 0.891 |