mlsanityData to 7 Oct 2026

TaSIL: Taylor Series Imitation Learning

arXiv 2205.14812 ↗First posted 30 May 2022cs.LGcited by 21

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

Abstract (arXiv)

We propose Taylor Series Imitation Learning (TaSIL), a simple augmentation to standard behavior cloning losses in the context of continuous control. TaSIL penalizes deviations in the higher-order Taylor series terms between the learned and expert policies. We show that experts satisfying a notion of textitincremental input-to-state stability are easy to learn, in the sense that a small TaSIL-augmented imitation loss over expert trajectories guarantees a small imitation loss over trajectories generated by the learned policy. We provide sample-complexity bounds for TaSIL that scale as tildeO(1/n) in the realizable setting, for n the number of expert demonstrations. Finally, we demonstrate experimentally the relationship between the robustness of the expert policy and the order of Taylor expansion required in TaSIL, and compare standard Behavior Cloning, DART, and DAgger with TaSIL-loss-augmented variants. In all cases, we show significant improvement over baselines across a variety of MuJoCo tasks.

Key facts. Cited by 21 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 17 of its 43 references that are in the map. First posted on arXiv 30 May 2022.

Builds on

17 of 43 references.

Closest ancestors · 4

Same direction, newest first

  1. 2021On the Sample Complexity of Stability Constrained Imitation Learning18 Feb
  2. 2020Imitation Learning with Stability and Safety Guarantees16 Dec
  3. 2018An Algorithmic Perspective on Imitation Learning16 Nov
  4. 2018Learning Contracting Vector Fields For Stable Imitation Learning13 Apr

Other references · 13

  1. 2016Gaussian Error Linear Units (GELUs)cited by 3,431
  2. 2016OpenAI Gymcited by 1,684
  3. 2020Knowledge Distillation: A Surveycited by 1,578
  4. 2016Generative Adversarial Imitation Learningcited by 1,576
  5. 2017End-to-end Driving via Conditional Imitation Learningcited by 479
  6. 2017DART: Noise Injection for Robust Imitation Learningcited by 187
  7. 2018Learning Without Mixing: Towards A Sharp Analysis of Linear System Identificationcited by 158
  8. 2017Sobolev Training for Neural Networkscited by 136
  9. 2018Knowledge Transfer with Jacobian Matchingcited by 73
  10. 2018Learning an Approximate Model Predictive Controller with Guaranteescited by 54
  11. 2020Learning Stability Certificates from Datacited by 45
  12. 2019Learning Stabilizable Nonlinear Dynamics with Contraction-Based Regularizationcited by 32
1 more
  1. 2020Regret Bounds for Adaptive Nonlinear Controlcited by 25

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 8; 2024: 3; 2025: 4; 2026: 3.

21 papers.

Sort:

By area:

  1. 2022Statistical Learning Theory for Control: A Finite Sample Perspectivecited by 52
  2. 2024JUICER: Data-Efficient Imitation Learning for Robotic Assemblycited by 36
  3. 2023Provable Guarantees for Generative Behavior Cloning: Bridging Low-Level Stability and High-Level Behaviorcited by 33
  4. 2022Multi-Task Imitation Learning for Linear Dynamical Systemscited by 22
  5. 2023Neural Operators of Backstepping Controller and Observer Gain Functions for Reaction-Diffusion PDEscited by 16
  6. 2023Butterfly Effects of SGD Noise: Error Amplification in Behavior Cloning and Autoregressioncited by 14
  7. 2024SAFE-GIL: SAFEty Guided Imitation Learning for Robotic Systemscited by 13
  8. 2023CCIL: Context-conditioned imitation learning for urban drivingcited by 12
  9. 2023Task Generalization with Stability Guarantees via Elastic Dynamical System Motion Policiescited by 10
  10. 2022Model Predictive Control via On-Policy Imitation Learningcited by 6
  11. 2023Oracle-Efficient Smoothed Online Learning for Piecewise Continuous Decision Makingcited by 5
  12. 2023The Power of Learned Locally Linear Models for Nonlinear Policy Optimizationcited by 3
9 more
  1. 2023On the Sample Complexity of Imitation Learning for Smoothed Model Predictive Controlcited by 3
  2. 2025A Test-Function Approach to Incremental Stabilitycited by 1
  3. 2025Distributionally Robust Imitation Learning: Layered Control Architecture for Certifiable Autonomycited by 1
  4. 2024Improved Sample Complexity of Imitation Learning for Barrier Model Predictive Controlcited by 0
  5. 2025A Model-Based Approach to Imitation Learning through Multi-Step Predictionscited by 0
  6. 2025Safety-Aware Imitation Learning via MPC-Guided Disturbance Injectioncited by 0
  7. 2026Understanding Behavior Cloning with Action Quantizationcited by 0
  8. 2026Accelerating trajectory optimization with Sobolev-trained diffusion policiescited by 0
  9. 2026Distributionally Robust and Safe Imitation Learningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025The Pitfalls of Imitation Learning when Actions are Continuous
  2. 2024Stable-BC: Controlling Covariate Shift with Stable Behavior Cloning
  3. 2024Is Behavior Cloning All You Need? Understanding Horizon in Imitation Learning
  4. 2022Data augmentation for efficient learning from parametric experts
  5. 2026Split the Differences, Pool the Rest: Provably Efficient Multi-Objective Imitation
  6. 2024Interactive and Hybrid Imitation Learning: Provably Beating Behavior Cloning
  7. 2026Difference-Aware Retrieval Policies for Imitation Learning
  8. 2023A Simple Solution for Offline Imitation from Observations and Examples with Possibly Incomplete Trajectories