mlsanityData to 14 Oct 2026

Towards Resolving Unidentifiability in Inverse Reinforcement Learning

arXiv 1601.06569 ↗First posted 25 Jan 2016cs.AIcited by 18

Not on the map. Why?

Abstract (arXiv)

We consider a setting for Inverse Reinforcement Learning (IRL) where the learner is extended with the ability to actively select multiple environments, observing an agent's behavior on each environment. We first demonstrate that if the learner can experiment with any transition dynamics on some fixed set of states and actions, then there exists an algorithm that reconstructs the agent's reward function to the fullest extent theoretically possible, and that requires only a small (logarithmic) number of experiments. We contrast this result to what is known about IRL in single fixed environments, namely that the true reward function is fundamentally unidentifiable. We then extend this setting to the more realistic case where the learner may not select any transition dynamic, but rather is restricted to some fixed set of environments that it may try. We connect the problem of maximizing the information derived from experiments to submodular function maximization and demonstrate that a greedy algorithm is near optimal (up to logarithmic factors). Finally, we empirically validate our algorithm on an environment inspired by behavioral psychology.

Key facts. Cited by 18 papers in the map (data through 14 Oct 2026). First posted on arXiv 25 Jan 2016.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2019: 1; 2020: 1; 2021: 2; 2022: 3; 2023: 3; 2024: 3; 2025: 5.

18 papers.

Sort:

By area:

  1. 2021Goal Misgeneralization in Deep Reinforcement Learningcited by 123
  2. 2021Identifiability in inverse reinforcement learningcited by 55
  3. 2025General agents contain world modelscited by 17
  4. 2020Efficient Exploration of Reward Functions in Inverse Reinforcement Learning via Bayesian Optimizationcited by 16
  5. 2022Identifiability and generalizability from multiple experts in Inverse Reinforcement Learningcited by 16
  6. 2023Reward-Consistent Dynamics Models are Strongly Generalizable for Offline Reinforcement Learningcited by 12
  7. 2023Inverse Decision Modeling: Learning Interpretable Representations of Behaviorcited by 10
  8. 2019Better-than-Demonstrator Imitation Learning via Automatically-Ranked Demonstrationscited by 6
  9. 2022Transferable Reward Learning by Dynamics-Agnostic Discriminator Ensemblecited by 5
  10. 2024How does Inverse RL Scale to Large State Spaces? A Provably Efficient Approachcited by 5
  11. 2024Inverse Reinforcement Learning with Sub-optimal Expertscited by 4
  12. 2025The Limits of Predicting Agents from Behaviourcited by 3
6 more
  1. 2023Eliciting Risk Aversion with Inverse Reinforcement Learning via Interactive Questioningcited by 2
  2. 2022Autonomous Assessment of Demonstration Sufficiency via Bayesian Inverse Reinforcement Learningcited by 1
  3. 2025Generalizing Behavior via Inverse Reinforcement Learning with Closed-Form Reward Centroidscited by 1
  4. 2025Learning from Failures: Understanding LLM Alignment through Failure-Aware Inverse RLcited by 1
  5. 2024Inference of Utilities and Time Preference in Sequential Decision-Makingcited by 0
  6. 2025Robustness in the Face of Partial Identifiability in Reward Learningcited by 0