mlsanityData to 7 Oct 2026

Domain-Invariant Per-Frame Feature Extraction for Cross-Domain Imitation Learning with Visual Observations

arXiv 2502.02867 ↗First posted 5 Feb 2025cs.CV cs.AI cs.LGcited by 0

On the map: Offline RL and imitation learning from data → Imitation learning methods

Abstract (arXiv)

Imitation learning (IL) enables agents to mimic expert behavior without reward signals but faces challenges in cross-domain scenarios with high-dimensional, noisy, and incomplete visual observations. To address this, we propose Domain-Invariant Per-Frame Feature Extraction for Imitation Learning (DIFF-IL), a novel IL method that extracts domain-invariant features from individual frames and adapts them into sequences to isolate and replicate expert behaviors. We also introduce a frame-wise time labeling technique to segment expert behaviors by timesteps and assign rewards aligned with temporal contexts, enhancing task performance. Experiments across diverse visual environments demonstrate the effectiveness of DIFF-IL in addressing complex visual tasks.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data. Builds on 13 of its 83 references that are in the map. First posted on arXiv 5 Feb 2025.

Builds on

13 of 83 references.

Closest ancestors · 6

Same direction, newest first

  1. 2022IL-flOw: Imitation Learning from Observation using Normalizing Flows19 May
  2. 2022Imitation Learning from Observations under Transition Model Disparity25 Apr
  3. 2021Domain-Robust Visual Imitation Learning with Mutual Information Constraints8 Mar
  4. 2020Primal Wasserstein Imitation Learning8 Jun
  5. 2018Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow1 Oct
  6. 2018Generative Adversarial Imitation from Observation17 Jul

Other references · 7

  1. 2018Soft Actor-Critic Algorithms and Applicationscited by 1,179
  2. 2018DeepMind Control Suitecited by 935
  3. 2015Empirical Evaluation of Rectified Activations in Convolutional Networkcited by 515
  4. 2017Learning Robust Rewards with Adversarial Inverse Reinforcement Learningcited by 485
  5. 2019Efficient Exploration via State Marginal Matchingcited by 186
  6. 2018OIL: Observational Imitation Learningcited by 11
  7. 2023Offline Imitation from Observation via Primal Wasserstein State Occupancy Matchingcited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2023Domain Adaptive Imitation Learning with Visual Observation
  2. 2023Robust Visual Imitation Learning with Inverse Dynamics Representations
  3. 2021Cross-domain Imitation from Observations
  4. 2021The Surprising Effectiveness of Representation Learning for Visual Imitation
  5. 2022Resolving Copycat Problems in Visual Imitation Learning via Residual Action Prediction
  6. 2023Visual Imitation Learning with Patch Rewards
  7. 2025Imitation Learning in the Deep Learning Era: A Novel Taxonomy and Recent Advances
  8. 2022Learn what matters: cross-domain imitation learning with task-relevant embeddings