Decoding Molecular Graph Embeddings with Reinforcement Learning
This work addresses a bottleneck in molecular graph generation for computational chemistry, though it appears incremental as it builds on existing graph-to-graph autoencoding methods.
The paper tackles the problem of decoding molecular graphs from latent embeddings by introducing RL-VAE, a graph-to-graph variational autoencoder that uses reinforcement learning, resulting in an efficient method that simplifies training and evaluation compared to previous approaches.
We present RL-VAE, a graph-to-graph variational autoencoder that uses reinforcement learning to decode molecular graphs from latent embeddings. Methods have been described previously for graph-to-graph autoencoding, but these approaches require sophisticated decoders that increase the complexity of training and evaluation (such as requiring parallel encoders and decoders or non-trivial graph matching). Here, we repurpose a simple graph generator to enable efficient decoding and generation of molecular graphs.