Improving Deep Learning Models in Drug Discovery

Term: 1/2019 – 10/2022 (36 months)

Partner: Merck KGaA

This project is concerned with the development of improvements to the current state-of-the-art Deep Learning models in Drug Discovery. Concretely, bioactivity and toxicity modeling, generative models, and image-based models are tackled with recent improvements such as attention mechanisms.