A Cascaded Supervised Learning Approach to Inverse Reinforcement Learning

By Edouard Klein, Bilal Piot, Matthieu Geist, Olivier Pietquin. 2013

58 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.

Citations in the corpus

Leveraging Demonstrations for Deep Reinforcement Learning on Robotics Problems with Sparse Rewards (2017, methodology): “To address continuous state spaces and unknown dynamics, [5] solve inverse RL by combining classification and regression.”
State Aware Imitation Learning (2017, methodology): “…[13] uses estimated feature expectations and linearity of the value function to derive the parameters of a linear reward function while the more recent Cascaded Supervised IRL (CSI) [14] derives the reward function by training a Support Vector Machine based on the observed temporal differences.”
Interpretable Batch IRL to Extract Clinician Goals in ICU Hypotension Management. (2020, background): “There exist other works 11 that optimize reward by setting action value function as a score metric of a multi-class classiﬁcation problem, an approach that still suffers from a linear off-policy evaluation problem and the tuning of problem-speciﬁc heuristics.”
Learning from Demonstration for Shaping through Inverse Reinforcement Learning (2016, background): “The agent attempts to maximize the long-term reward determined by the reward and transition functions that are initially unknown to the agent.”
Primal Wasserstein Imitation Learning (2020, background): “Existing approaches derive a reward function without interaction with the environment (Boularias et al., 2011; Klein et al., 2013; 2012; Piot et al., 2016).”
A Survey of Inverse Reinforcement Learning: Challenges, Methods and Progress (2018, methodology): “An extension of the above method [41] (called csi) estimates the transition probabilities if they are unknown.”