Generating stable molecules using imitation and reinforcement learning.
Søren Ager MeldgaardJonas KöhlerHenrik Lund MortensenMads-Peter V. ChristiansenFrank NoéBjørk HammerPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- function approximation
- temporal difference
- state space
- reinforcement learning algorithms
- neural network
- imitation learning
- markov decision processes
- robotic control
- chemical compounds
- learning agents
- generation process
- model free
- optimal control
- optimal policy
- multi agent
- learning algorithm
- learning problems
- action selection
- transfer learning
- dynamic programming
- temporal difference learning
- case study
- policy search
- machine learning