Augmenting Genetic Algorithms with Deep Neural Networks for Exploring the Chemical Space

Abstract

Challenges in natural sciences can often be phrased as optimization problems. Machine learning techniques have recently been applied to solve such problems. One example in chemistry is the design of tailor-made organic materials and molecules, which requires efficient methods to explore the chemical space. We present a genetic algorithm (GA) that is enhanced with a neural network (DNN) based discriminator model to improve the diversity of generated molecules and at the same time steer the GA. We show that our algorithm outperforms other generative models in optimization tasks. We furthermore present a way to increase interpretability of genetic algorithms, which helped us to derive design principles.

arXiv comment9+3 Pages, 7+4 figures, 2 tables. Comments are welcome! (code is available at: https://github.com/aspuru-guzik-group/GA)

References in corpus (7)

• Auto-Encoding Variational Bayes
• Junction Tree Variational Autoencoder for Molecular Graph Generation
• Objective-Reinforced Generative Adversarial Networks (ORGAN) for Sequence Generation Models
• Constrained Graph Variational Autoencoders for Molecule Design
• ChemTS: An Efficient Python Library for de novo Molecular Generation
• Learning Multimodal Graph-to-Graph Translation for Molecular Optimization
• DEFactor: Differentiable Edge Factorization-based Probabilistic Graph Generation

Cited by in corpus (11)

• Review: Deep Learning in Electron Microscopy
• A Deep Generative Model for Molecule Optimization via One Fragment Modification
• Computer-inspired Quantum Experiments
• Generative Chemical Transformer: Neural Machine Learning of Molecular Geometric Structures from Chemical Language via Attention
• Learning To Navigate The Synthetically Accessible Chemical Space Using Reinforcement Learning
• Deep Molecular Dreaming: Inverse machine learning for de-novo molecular design and interpretability with surjective representations
• Golem: An algorithm for robust experiment and process optimization
• Learning the Travelling Salesperson Problem Requires Rethinking Generalization
• Curiosity in exploring chemical space: Intrinsic rewards for deep molecular reinforcement learning
• Amortized Tree Generation for Bottom-up Synthesis Planning and Synthesizable Molecular Design
• An In-depth Summary of Recent Artificial Intelligence Applications in Drug Design