Towards "AlphaChem": Chemical Synthesis Planning with Tree Search and Deep Neural Network Policies
arXiv:1702.00020
Abstract
Retrosynthesis is a technique to plan the chemical synthesis of organic molecules, for example drugs, agro- and fine chemicals. In retrosynthesis, a search tree is built by analysing molecules recursively and dissecting them into simpler molecular building blocks until one obtains a set of known building blocks. The search space is intractably large, and it is difficult to determine the value of retrosynthetic positions. Here, we propose to model retrosynthesis as a Markov Decision Process. In combination with a Deep Neural Network policy learned from essentially the complete published knowledge of chemistry, Monte Carlo Tree Search (MCTS) can be used to evaluate positions. In exploratory studies, we demonstrate that MCTS with neural network policies outperforms the traditionally used best-first search with hand-coded heuristics.
4 pages, 1 figure
References in corpus (1)
Cited by in corpus (6)
- Alchemy: A Quantum Chemistry Dataset for Benchmarking AI Models
- Chess AI: Competing Paradigms for Machine Intelligence
- Pipelines for Procedural Information Extraction from Scientific Literature: Towards Recipes using Machine Learning and Data Science
- Barking up the right tree: an approach to search over molecule synthesis DAGs
- World Programs for Model-Based Learning and Planning in Compositional State and Action Spaces
- Scientists in silico?