Rapidly exploring learning trees
Inverse Reinforcement Learning (IRL) for path planning enables robots to learn cost functions for difficult tasks from demonstration, instead of hard-coding them. However, IRL methods face practical limitations that stem from the need to repeat costly planning procedures. In this paper, we propose R...
Autori principali: | Shiarlis, K, Messias, J, Whiteson, S |
---|---|
Natura: | Conference item |
Pubblicazione: |
IEEE
2017
|
Documenti analoghi
-
Inverse reinforcement learning from failure
di: Shiarlis, K, et al.
Pubblicazione: (2016) -
TACO: Learning task decomposition via temporal alignment for control
di: Shiarlis, K, et al.
Pubblicazione: (2018) -
Dynamic-depth context tree weighting
di: Messias, J, et al.
Pubblicazione: (2018) -
Learning from demonstration in the wild
di: Behbahani, F, et al.
Pubblicazione: (2019) -
VariBAD: a very good method for Bayes-adaptive deep RL via meta-learning
di: Zintgraf, L, et al.
Pubblicazione: (2020)