Active Reward Learning from Critiques

80 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

End-to-End Robotic Reinforcement Learning without Reward Engineering (2019, background): “Recent work on active learning for inverse RL has sought to reduce the required number of demonstrations [2, 6, 26, 5], but”
Learning Human Objectives by Evaluating Hypothetical Behavior (2019, background): “In step (2) we represent ˆ R by classifying state transitions as good, unsafe, or neutral – similar to Cui & Niekum (2018) – and assigning a known, constant reward to each of these three categories: ˆ R ( s, a, s (cid:48) ) = c ∈{ good , unsafe , neutral } p θ ( c | s, a, s (cid:48) ) R c , (2)…”
A Review of Robot Learning for Manipulation: Challenges, Representations, and Algorithms (2019, methodology): “Most notably, the IRL problem is fundamentally ill-posed—infinitely many reward functions exist that result in the same optimal policy [275].”
The EMPATHIC Framework for Task Learning from Implicit Human Feedback (2020, background): “We view learning from implicit human feedback (LIHF) as complementary to learning from explicit human teaching, which might take the form of demonstrations [1], evaluative feedback [2, 3], or other communicative modalities [4, 5, 6, 7, 8].”
Risk-Aware Active Inverse Reinforcement Learning (2018, methodology): “Many existing methods [7, 5] are purely exploratory since they reason only about statistical measures instead of the performance of a specific policy when selecting active queries.”