An Exploration Strategy Improves the Diversity of De Novo Ligands Using Deep Reinforcement Learning – A Case for the 3 Adenosine A2A Receptor

doi 10.26434/chemrxiv.7436789