Off-Policy Evaluation of the Performance of a Robot Swarm: Importance Sampling to Assess Potential Modifications to the Finite-State Machine That Controls the Robots.
Federico PagnozziMauro BirattariPublished in: Frontiers Robotics AI (2021)
Keyphrases
- importance sampling
- finite state machines
- policy evaluation
- monte carlo
- mobile robot
- variance reduction
- markov chain
- temporal difference
- model checking
- least squares
- particle filter
- kalman filter
- markov chain monte carlo
- hidden markov models
- model free
- function approximation
- reinforcement learning
- approximate inference