AlwaysSafe: Reinforcement Learning without Safety Constraint Violations during Training.

Thiago D. Simão Nils Jansen Matthijs T. J. Spaan

Published in: AAMAS (2021)

Keyphrases

constraint violations
reinforcement learning
special case
temporal constraints
state space
neural network
genetic algorithm
image sequences
search algorithm
simulated annealing
evolutionary computation
multiple criteria