Symmetry-Aware Actor-Critic for 3D Molecular Design

11/25/2020
by   Gregor N. C. Simm, et al.
0

Automating molecular design using deep reinforcement learning (RL) has the potential to greatly accelerate the search for novel materials. Despite recent progress on leveraging graph representations to design molecules, such methods are fundamentally limited by the lack of three-dimensional (3D) information. In light of this, we propose a novel actor-critic architecture for 3D molecular design that can generate molecular structures unattainable with previous approaches. This is achieved by exploiting the symmetries of the design process through a rotationally covariant state-action representation based on a spherical harmonics series expansion. We demonstrate the benefits of our approach on several 3D molecular design tasks, where we find that building in such symmetries significantly improves generalization and the quality of generated molecules.

READ FULL TEXT

page 8

page 9

page 17

research
03/10/2023

Understanding the Synergies between Quality-Diversity and Deep Reinforcement Learning

The synergies between Quality-Diversity (QD) and Deep Reinforcement Lear...
research
02/01/2022

Scalable Fragment-Based 3D Molecular Design with Reinforcement Learning

Machine learning has the potential to automate molecular design and dras...
research
02/18/2020

Reinforcement Learning for Molecular Design Guided by Quantum Mechanics

Automating molecular design using deep reinforcement learning (RL) holds...
research
12/17/2021

Symmetry-aware Neural Architecture for Embodied Visual Navigation

Visual exploration is a task that seeks to visit all the navigable areas...
research
09/06/2023

Addressing Imperfect Symmetry: a Novel Symmetry-Learning Actor-Critic Extension

Symmetry, a fundamental concept to understand our environment, often ove...
research
07/11/2014

SketchBio: A Scientist's 3D Interface for Molecular Modeling and Animation

Background: Because of the difficulties involved in learning and using 3...
research
02/25/2021

CPG-ACTOR: Reinforcement Learning for Central Pattern Generators

Central Pattern Generators (CPGs) have several properties desirable for ...

Please sign up or login with your details

Forgot password? Click here to reset