mlsanityData to 7 Oct 2026

ConTraIRL: Factorized Contrastive Abstractions for Transferable IRL

arXiv 2606.03017 ↗First posted 2 Jun 2026cs.LG cs.AI cs.ROcited by 0

On the map: Offline RL and imitation learning from data → Inverse reinforcement learning

Abstract (arXiv)

Reward transfer in Inverse Reinforcement Learning (IRL) is unreliable when policies must generalize to unseen combinations of environment dynamics and task goals. We propose Factorized Contrastive Abstractions for Transferable IRL (ConTraIRL), a framework that enables compositional reward transfer by learning decoupled latent representations of these two factors. ConTraIRL uses a dual-encoder architecture that maps observations into separate dynamics and goal latent spaces, trained with a dual contrastive objective. Temporal alignment encourages the dynamics encoder to learn goal-invariant structure, while the goal encoder captures dynamics-invariant features. This factorization supports reward inference under recombined dynamics-goal settings. Experiments on continuous control benchmarks demonstrate effective few-shot transfer to unseen dynamics-goal pairings, improving sample efficiency and reward recovery over transfer IRL baselines.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Inverse reinforcement learning, one of 13 directions in Offline RL and imitation learning from data. Builds on 30 of its 35 references that are in the map. First posted on arXiv 2 Jun 2026.

Builds on

30 of 35 references.

Closest ancestors · 8

Same direction, newest first

  1. 2026CoMI-IRL: Contrastive Multi-Intention Inverse Reinforcement Learning7 Feb
  2. 2025TW-CRL: Time-Weighted Contrastive Reward Learning for Efficient Inverse Reinforcement Learning8 Apr
  3. 2025Inversely Learning Transferable Rewards via Abstracted States3 Jan
  4. 2024Non-Adversarial Inverse Reinforcement Learning via Successor Feature Matching11 Nov
  5. 2023Multi-task Hierarchical Adversarial Inverse Reinforcement Learning22 May
  6. 2022Environment Design for Inverse Reinforcement Learning26 Oct
  7. 2019Meta-Inverse Reinforcement Learning with Probabilistic Context Variables20 Sep
  8. 2018A Survey of Inverse Reinforcement Learning: Challenges, Methods and Progress18 Jun

Other references · 22

  1. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  2. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  3. 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,841
  4. 2020Supervised Contrastive Learningcited by 2,803
  5. 2016Generative Adversarial Imitation Learningcited by 1,576
  6. 2016HyperNetworkscited by 1,003
  7. 2020What Makes for Good Views for Contrastive Learning?cited by 843
  8. 2017Learning Robust Rewards with Adversarial Inverse Reinforcement Learningcited by 485
  9. 2016Successor Features for Transfer in Reinforcement Learningcited by 331
  10. 2015Contextual Markov Decision Processescited by 213
  11. 2021IQ-Learn: Inverse soft-Q Learning for Imitationcited by 191
  12. 2020Adversarial Self-Supervised Contrastive Learningcited by 138
10 more
  1. 2019Transfer in Deep Reinforcement Learning Using Successor Features and Generalised Policy Improvementcited by 88
  2. 2022Graph Inverse Reinforcement Learning from Diverse Videoscited by 45
  3. 2019Assistive Gym: A Physics Simulation Framework for Assistive Roboticscited by 44
  4. 2020State-only Imitation with Transition Dynamics Mismatchcited by 37
  5. 2022Robust Imitation Learning against Variations in Environment Dynamicscited by 26
  6. 2021PsiPhi-Learning: Reinforcement Learning with Demonstrations using Successor Features and Inverse Temporal Difference Learningcited by 23
  7. 2020Joint Goal and Strategy Inference across Heterogeneous Demonstrators via Reward Network Distillationcited by 19
  8. 2024When does compositional structure yield compositional generalization? A kernel theorycited by 19
  9. 2022Fast Lifelong Adaptive Inverse Reinforcement Learning from Demonstrationscited by 14
  10. 2021A New Representation of Successor Features for Transfer across Dissimilar Environmentscited by 13

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Inverse Reinforcement Learning without an Optimal Demonstrator: A Feasible Reward Set Approach
  2. 2024Towards the Transferability of Rewards Recovered via Regularized Inverse Reinforcement Learning
  3. 2026Generalize and Guide: Decomposing Rewards for Few-Shot Inverse Reinforcement Learning
  4. 2026Scalable Multi-Task Inverse Reinforcement Learning
  5. 2025Consistent Zero-Shot Imitation with Contrastive Goal Inference
  6. 2026IR3: Contrastive Inverse Reinforcement Learning for Interpretable Detection and Mitigation of Reward Hacking
  7. 2021Versatile Inverse Reinforcement Learning via Cumulative Rewards
  8. 2025Reward Compatibility: A Framework for Inverse RL