A model-free RL algorithm learns policies by sequentially maximizing safety probability, then LTL satisfaction probability, then discounted QoC rewards in unknown MDPs.
Kress-Gazit, F
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2021 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Model-Free Learning of Safe yet Effective Controllers
A model-free RL algorithm learns policies by sequentially maximizing safety probability, then LTL satisfaction probability, then discounted QoC rewards in unknown MDPs.