Open Peer Review. Open Publishing. Open Access. Open Discussion. Open Directory. Open Recommendations. Open API. Open Source.
Using Deep Reinforcement Learning to Generate Rationales for Molecules
Benson Chen, Connor Coley, Regina Barzilay, Tommi Jaakkola
Feb 15, 2018 (modified: Feb 15, 2018)ICLR 2018 Conference Blind Submissionreaders: everyoneShow Bibtex
Abstract:Deep learning algorithms are increasingly used in modeling chemical processes. However, black box predictions without rationales have limited used in practical applications, such as drug design. To this end, we learn to identify molecular substructures -- rationales -- that are associated with the target chemical property (e.g., toxicity). The rationales are learned in an unsupervised fashion, requiring no additional information beyond the end-to-end task. We formulate this problem as a reinforcement learning problem over the molecular graph, parametrized by two convolution networks corresponding to the rationale selection and prediction based on it, where the latter induces the reward function. We evaluate the approach on two benchmark toxicity datasets. We demonstrate that our model sustains high performance under the additional constraint that predictions strictly follow the rationales. Additionally, we validate the extracted rationales through comparison against those described in chemical literature and through synthetic experiments.
TL;DR:We use a reinforcement learning over molecular graphs to generate rationales for interpretable molecular property prediction.