mlsanityData to 7 Oct 2026

Adversarial Imitation Learning with Trajectorial Augmentation and Correction

arXiv 2103.13887 ↗First posted 25 Mar 2021cs.LGcited by 11

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

Abstract (arXiv)

Deep Imitation Learning requires a large number of expert demonstrations, which are not always easy to obtain, especially for complex tasks. A way to overcome this shortage of labels is through data augmentation. However, this cannot be easily applied to control tasks due to the sequential nature of the problem. In this work, we introduce a novel augmentation method which preserves the success of the augmented trajectories. To achieve this, we introduce a semi-supervised correction network that aims to correct distorted expert actions. To adequately test the abilities of the correction network, we develop an adversarial data augmented imitation architecture to train an imitation agent using synthetic experts. Additionally, we introduce a metric to measure diversity in trajectory datasets. Experiments show that our data augmentation strategy can improve accuracy and convergence time of adversarial imitation while preserving the diversity between the generated and real trajectories.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data. Builds on 30 of its 45 references that are in the map. First posted on arXiv 25 Mar 2021.

Builds on

30 of 45 references.

Closest ancestors · 1

Same direction, newest first

  1. 2018Multi-Agent Generative Adversarial Imitation Learning26 Jul

Other references · 29

  1. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  2. 2015Continuous control with deep reinforcement learningcited by 3,279
  3. 2015Trust Region Policy Optimizationcited by 2,503
  4. 2016OpenAI Gymcited by 1,684
  5. 2016Understanding deep learning requires rethinking generalizationcited by 1,657
  6. 2016Generative Adversarial Imitation Learningcited by 1,576
  7. 2017Random Erasing Data Augmentationcited by 1,235
  8. 2016End to End Learning for Self-Driving Carscited by 1,000
  9. 2017Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrationscited by 685
  10. 2017Deep Mutual Learningcited by 621
  11. 2017Zero-Shot Learning -- The Good, the Bad and the Uglycited by 254
  12. 2017One-Shot Imitation Learningcited by 232
17 more
  1. 2017One-Shot Visual Imitation Learning via Meta-Learningcited by 225
  2. 2018Low-Shot Learning from Imaginary Datacited by 181
  3. 2015Learning Continuous Control Policies by Stochastic Value Gradientscited by 156
  4. 2017Learning From Noisy Large-Scale Datasets With Minimal Supervisioncited by 115
  5. 2018GAN Augmentation: Augmenting Training Data using Generative Adversarial Networkscited by 107
  6. 2017Vision-Based Multi-Task Manipulation for Inexpensive Robots Using End-To-End Learning from Demonstrationcited by 100
  7. 2017Robust Imitation of Diverse Behaviorscited by 85
  8. 2017DeLiGAN : Generative Adversarial Networks for Diverse and Limited Datacited by 77
  9. 2019Small Data Challenges in Big Data Era: A Survey of Recent Progress on Unsupervised and Semi-Supervised Methodscited by 62
  10. 2019Data Augmentation Using GANscited by 58
  11. 2017Zero-Shot Deep Domain Adaptationcited by 35
  12. 2018Adversarial Teacher-Student Learning for Unsupervised Domain Adaptationcited by 27
  13. 2019AC-Teach: A Bayesian Actor-Critic Method for Policy Learning with an Ensemble of Suboptimal Teacherscited by 27
  14. 2018Task-Oriented Hand Motion Retargeting for Dexterous Manipulation Imitationcited by 25
  15. 2019Conditional Vehicle Trajectories Prediction in CARLA Urban Environmentcited by 6
  16. 2018Imitation Learning for End to End Vehicle Longitudinal Control with Forward Cameracited by 5
  17. 2017Exact Mean Computation in Dynamic Time Warping Spacescited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 3; 2023: 2; 2024: 1; 2025: 1; 2026: 3.

11 papers.

  1. 2025Dexterous Manipulation through Imitation Learning: A Surveycited by 18
  2. 2021Demonstration-Efficient Guided Policy Search via Imitation of Robust Tube MPCcited by 10
  3. 2023Offline Imitation Learning with Variational Counterfactual Reasoningcited by 7
  4. 2024Fusion Dynamical Systems with Machine Learning in Imitation Learning: A Comprehensive Overviewcited by 3
  5. 2022Modular Adaptive Policy Selection for Multi-Task Imitation Learning through Task Divisioncited by 2
  6. 2022Output Feedback Tube MPC-Guided Data Augmentation for Robust, Efficient Sensorimotor Policy Learningcited by 2
  7. 2022Evaluating the Effectiveness of Corrective Demonstrations and a Low-Cost Sensor for Dexterous Manipulationcited by 0
  8. 2023Simple Noisy Environment Augmentation for Reinforcement Learningcited by 0
  9. 2026PhysHanDI: Physics-Based Reconstruction of Hand-Deformable Object Interactionscited by 0
  10. 2026Dynamic Full-body Motion Agent with Object Interaction via Blending Pre-trained Modular Controllerscited by 0
  11. 2026A Systematic Approach for Selecting Trajectories for Data Augmentationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Unlabeled Imperfect Demonstrations in Adversarial Imitation Learning
  2. 2023Sample-efficient Adversarial Imitation Learning
  3. 2021A Pragmatic Look at Deep Imitation Learning
  4. 2019Sample-efficient Adversarial Imitation Learning from Observation
  5. 2022Latent Policies for Adversarial Imitation Learning
  6. 2023Improving Generalization in Game Agents with Data Augmentation in Imitation Learning
  7. 2018Sample-Efficient Imitation Learning via Generative Adversarial Nets
  8. 2026Latent Wasserstein Adversarial Imitation Learning