Nonasymptotic analysis shows sub-Gaussian action errors in behavior cloning propagate through gain-dependent closed-loop dynamics to produce sub-Gaussian position errors whose tail is governed by a proxy matrix and amplification index that depends on controller stiffness and damping.
Pd control with desired gravity compensation of robotic manipulators: a review
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
fields
cs.RO 2years
2026 2roles
background 2polarities
background 2representative citing papers
Controller gains affect learnability differently for behavior cloning, RL from scratch, and sim-to-real transfer, so optimal gains depend on the learning paradigm rather than desired task behavior.
citing papers explorer
-
A Nonasymptotic Theory of Gain-Dependent Error Dynamics in Behavior Cloning
Nonasymptotic analysis shows sub-Gaussian action errors in behavior cloning propagate through gain-dependent closed-loop dynamics to produce sub-Gaussian position errors whose tail is governed by a proxy matrix and amplification index that depends on controller stiffness and damping.
-
Tune to Learn: How Controller Gains Shape Robot Policy Learning
Controller gains affect learnability differently for behavior cloning, RL from scratch, and sim-to-real transfer, so optimal gains depend on the learning paradigm rather than desired task behavior.