Reducing Safety Interventions in Provably Safe Reinforcement Learning.

Jakob Thumm Guillaume Pelat Matthias Althoff

Published in: IROS (2023)

Keyphrases

reinforcement learning
function approximation
state space
learning algorithm
robotic control
data sets

action selection
information systems
information retrieval
learning process
evolutionary algorithm

markov decision processes
data mining
optimal control
real world
temporal difference

temporal difference learning
stochastic approximation
autonomous learning
driving behavior
neural network