Exploring Deep Recurrent Models with Reinforcement Learning for Molecule Design

By Daniel Neil, Marwin H. S. Segler, L. Guasch, Mohamed Ahmed, Dean Plumbley, M. Sellwood, Nathan Brown. 2018

79 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Marc Brockschmidt
2. Miltiadis Allamanis
3. Chao Yu
4. Nathan Brown
5. Qi Liu
6. Uri Alon
7. Chence Shi
8. Jiming Liu

Citations in the corpus

• Reinforcement Learning in Healthcare: A Survey (2019, methodology): “[163] investigated the PPO method in molecular generation.”
• Model-based reinforcement learning for biological sequence design (2020, unknown): “For example, hillclimb-MLE (Neil et al., 2018) performs maximum-likelihood training on the top k sequences from step (i).”
• A graph-based genetic algorithm and generative model/Monte Carlo tree search for the exploration of chemical space (2018, result): “;2;3;4;5;6;7;8] While the performance of these generative models have been impressive, they are usually not compared to more traditional approaches such as genetic algorithms (GAs), [9;10;11;12] and recent work by Tsuda and co-workers [13] suggest that non-ML approaches may be competitive.”
• Structure-aware generation of drug-like molecules (2021, methodology): “Early generative methods designed SMILES [7], string-based molecular representations, using language models, RNNs, and GANs [8, 9, 10, 11, 12], discarding molecular graph information.”
• GuacaMol: Benchmarking Models for De Novo Molecular Design (2018, methodology): “49 The model used was an LSTM with 3 layers of hidden size of 1024.”
• GraphAF: a Flow-based Autoregressive Model for Molecular Graph Generation (2020, background): “2 RELATED WORK A variety of deep generative models have been proposed for molecular graph generation recently (Segler et al., 2017; Olivecrona et al., 2017; Samanta et al., 2018; Neil et al., 2018).”
• Constrained Graph Variational Autoencoders for Molecule Design (2018, methodology): “More specifically, generating and optimizing chemical molecules has been identified as an important real-world application for this set of techniques [8, 23, 24, 28, 29].”
• Multi-objective de novo drug design with conditional graph generative model (2018, methodology): “[27] created a benchmark analysis of various RL based method in different tasks of molecule design.”