Tuning the Molecular Weight Distribution from Atom Transfer Radical Polymerization Using Deep Reinforcement Learning

  title={Tuning the Molecular Weight Distribution from Atom Transfer Radical Polymerization Using Deep Reinforcement Learning},
  author={Haicheng Li and Christopher R. Collins and Thomas G Ribelli and K. Matyjaszewski and Geoffrey J. Gordon and T. Kowalewski and D. Yaron},
  journal={arXiv: Chemical Physics},
We devise a novel technique to control the shape of polymer molecular weight distributions (MWDs) in atom transfer radical polymerization (ATRP). This technique makes use of recent advances in both simulation-based, model-free reinforcement learning (RL) and the numerical simulation of ATRP. A simulation of ATRP is built that allows an RL controller to add chemical reagents throughout the course of the reaction. The RL controller incorporates fully-connected and convolutional neural network… Expand

Tables from this paper

Deep Learning for Deep Chemistry: Optimizing the Prediction of Chemical Patterns
In this review, the most exciting developments concerning the use of ML in a range of different chemical scenarios are described, evidencing the breadth of potential applications of these emerging multidimensional approaches. Expand
Autonomous reinforcement learning agent for chemical vapor deposition synthesis of quantum materials
Predictive materials synthesis is the primary bottleneck in realizing functional and quantum materials. Strategies for synthesis of promising materials are currently identified by time-consumingExpand
Deep learning for molecular generation and optimization - a review of the state of the art
Several important high level themes emerge, including the shift away from the SMILES string representation of molecules towards more sophisticated representations such as graph grammars and 3D representations, the importance of reward function design, the need for better standards for benchmarking and testing, and the benefits of adversarial training and reinforcement learning over maximum likelihood based training. Expand
Artificial neural networks towards average properties targets in styrene ARGET-ATRP
Abstract The application of artificial neural networks (ANN) in the synthesis of polystyrene via ARGET-ATRP was presented for the first time. In this research, it was utilized a deterministicExpand
Machine-learning-based simulation and fed-batch control of cyanobacterial-phycocyanin production in Plectonema by artificial neural network and deep reinforcement learning
This application utilizes the Asynchronous Advantage Actor-Critic (A3C) algorithm, a member of the DRL family, that takes advantage of actor-critic algorithm and asynchronous learning by parallel learning agents to achieve stability and efficiency of the learning process. Expand
Predictive design of polymer molecular weight distributions in anionic polymerization
Molecular weight distributions (MWD) have a substantial impact on a diverse set of polymer physical and rheological properties, from processability and stiffness to many aspects of block copolymerExpand
Controlling polymer properties through the shape of the molecular-weight distribution
The manipulation of a polymer’s properties without altering its chemical composition is a major challenge in polymer chemistry, materials science and engineering. Although variables such as chemicalExpand
A predictive framework for mixing low dispersity polymer samples to design custom molecular weight distributions
The physical properties of polymer samples are dependent on the overall shape and breadth of the molecular weight distribution (MWD). A small number of methods are available to tune the shape andExpand
Potentials and challenges of polymer informatics: exploiting machine learning for polymer design
This study reviews and discusses the applications of machine learning on different aspects of the polymer design process through four perspectives: polymer databases, representation (descriptor) of polymers, predictive models for polymer properties, and polymer design strategy. Expand
Recent Developments and Future Challenges in Controlled Radical Polymerization: A 2020 Update
Summary This perspective summarizes the latest exciting developments in controlled radical polymerization during the last decade (2010–2020). Our focus is to critically highlight strengths andExpand


Deep reinforcement learning for de novo drug design
The ReLeaSE method is used to design chemical libraries with a bias toward structural complexity or toward compounds with maximal, minimal, or specific range of physical properties, such as melting point or hydrophobicity. Expand
Optimizing Chemical Reactions with Deep Reinforcement Learning
This model outperformed a state-of-the-art blackbox optimization algorithm by using 71% fewer steps on both simulations and real reactions, and showed a better performance after training on reactions with similar or even dissimilar underlying mechanisms, which demonstrates its learning ability. Expand
Molecular de-novo design through deep reinforcement learning
A method to tune a sequence-based generative model for molecular de novo design that through augmented episodic likelihood can learn to generate structures with certain specified desirable properties is introduced. Expand
Automatic Control of Polymer Molecular Weight during Synthesis
The continuous record of monomer and polymer concentrations, Cm and Cp, and cumulative weight-average mass, Mw, furnished by automatic continuous online monitoring of polymerization reactions (ACOMP)Expand
Deeply AggreVaTeD: Differentiable Imitation Learning for Sequential Prediction
This work presents two gradient procedures that can learn neural network policies for several problems, including a sequential prediction task and several high-dimensional robotics control problems and provides a comprehensive theoretical study of IL. Expand
Benchmarking Deep Reinforcement Learning for Continuous Control
This work presents a benchmark suite of continuous control tasks, including classic tasks like cart-pole swing-up, tasks with very high state and action dimensionality such as 3D humanoid locomotion, task with partial observations, and tasks with hierarchical structure. Expand
A Brief Survey of Deep Reinforcement Learning
This survey will cover central algorithms in deep reinforcement learning, including the deep Q-network, trust region policy optimisation, and asynchronous advantage actor-critic, and highlight the unique advantages of deep neural networks, focusing on visual understanding via reinforcement learning. Expand
Human-level control through deep reinforcement learning
This work bridges the divide between high-dimensional sensory inputs and actions, resulting in the first artificial agent that is capable of learning to excel at a diverse array of challenging tasks. Expand
Evolving deep unsupervised convolutional networks for vision-based reinforcement learning
Both the MPCNN preprocessor and the RNN controller are evolved successfully to control a car in the TORCS racing simulator using only visual input, the first use of deep learning in the context evolutionary RL. Expand
Continuous control with deep reinforcement learning
This work presents an actor-critic, model-free algorithm based on the deterministic policy gradient that can operate over continuous action spaces, and demonstrates that for many of the tasks the algorithm can learn policies end-to-end: directly from raw pixel inputs. Expand