A contractive dynamical-system policy, built from recurrent equilibrium networks and coupling layers, that guarantees out-of-sample recovery and is shown to beat stable baselines on imitation benchmarks.
Neural dynamic policies for end-to-end sensorimotor learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Contractive Dynamical Imitation Policies for Efficient Out-of-Sample Recovery
A contractive dynamical-system policy, built from recurrent equilibrium networks and coupling layers, that guarantees out-of-sample recovery and is shown to beat stable baselines on imitation benchmarks.