Learning Robust Rewards with Adverserial Inverse Reinforcement Learning

By Justin Fu, Katie Luo, Sergey Levine. 2018

257 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

Acme: A Research Framework for Distributed Reinforcement Learning (2020, background): “, 2017) or using different formulation of the rewards (Fu et al., 2018).”
A Divergence Minimization Perspective on Imitation Learning Methods (2019, methodology): “We begin by presenting f -MAX, a generalization of AIRL [1] which provides a more intuitive interpretation of what similar algorithms accomplish.”
Efficient Exploration via State Marginal Matching (2019, methodology): “Similar to some inverse RL algorithms (Ho & Ermon, 2016; Fu et al., 2018), our method iterates between learning a policy and learning a reward function, though our reward function is obtained via a density model instead of a discriminator.”
Goal-Conditioned Imitation Learning using Score-based Diffusion Policies (2023, background): “One popular approach is the use of Generative Adversarial Networks (GANs) [13, 11].”
Positive-Unlabeled Reward Learning (2019, unknown): “We take a state-only formulation of GAIL that is known to improve the robustness of the algorithm [9, 27] and also does not require expert action information.”
Fighting Failures with FIRE: Failure Identification to Reduce Expert Burden in Intervention-Based Learning (2020, background): “These demonstrations do not necessarily achieve maximal reward under the environment constraints due to varying degrees of sub-optimality in each individual expert action.”
Multi-task Maximum Entropy Inverse Reinforcement Learning (2018, methodology): “Both of these shortcomings are addressed by guided cost learning (Finn et al., 2016) and its successor adversarial IRL (Fu et al., 2018), scalable approximations of MCE IRL.”
Generative predecessor models for sample-efficient imitation learning (2019, background): “The arguably most successful approach to IRL aims to match the state-action distribution to that of the demonstrations (Ziebart et al., 2008) with recent approaches extending these ideas to the model free case (Boularias et al., 2011; Finn et al., 2016; Fu et al., 2018).”
Learning a Prior over Intent via Meta-Inverse Reinforcement Learning (2018, methodology): “, 2017a), which uses importance sampling and regularization for scalability to highdimensional spaces, and adversarial IRL (Fu et al., 2018).”
Example-Driven Model-Based Reinforcement Learning for Solving Long-Horizon Visuomotor Tasks (2021, background): “Motivated by this, many works have studied the reward speciﬁcation, including inverse reinforcement learning [38] with robot demonstrations [39, 40, 41, 42, 43], learning rewards from user preferences [44, 45, 46, 47], and learning rewards from videos of humans [48, 49].”
Imitation Learning from Video by Leveraging Proprioception (2019, methodology): “, 2019] and improving reward representation [Fu et al., 2018; Qureshi et al., 2019].”
RvS: What is Essential for Offline RL via Supervised Learning? (2021, background): “The best-performing architectures are generally larger than the architectures used in standard online RL [28] and imitation learning [11, 22].”