Causal versus Marginal Shapley Values for Robotic Lever Manipulation Controlled using Deep Reinforcement Learning.
Sindre Benjamin RemmanInga StrümkeAnastasios M. LekkasPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- manipulation tasks
- real time
- bayesian networks
- function approximation
- mobile robot
- real robot
- probability distribution
- state space
- attribute values
- standard deviation
- parameter values
- causal relationships
- markov decision processes
- causal reasoning
- robotic systems
- dynamic programming
- neural network