# Smooth Imitation Learning for Online Sequence Prediction

arXiv: https://arxiv.org/abs/1606.00968 · HTML: https://mlsanity.com/paper/1606.00968/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Jun 2016.

## Built on by (most cited first, 9 of 9)

- [Strictly Batch Imitation Learning by Energy-based Distribution Matching](https://mlsanity.com/paper/2006.14154/) (2020)
- [Topo-boundary: A Benchmark Dataset on Topological Road-boundary Detection Using Aerial Images for Autonomous Driving](https://mlsanity.com/paper/2103.17119/) (2021)
- [Structured Agent Distillation for Large Language Model](https://mlsanity.com/paper/2505.13820/) (2025)
- [Follow Your Path: a Progressive Method for Knowledge Distillation](https://mlsanity.com/paper/2107.09305/) (2021)
- [On Efficient Online Imitation Learning via Classification](https://mlsanity.com/paper/2209.12868/) (2022)
- [Learning-Augmented Decentralized Online Convex Optimization in Networks](https://mlsanity.com/paper/2306.10158/) (2023)
- [Distributionally Robust Imitation Learning: Layered Control Architecture for Certifiable Autonomy](https://mlsanity.com/paper/2512.17899/) (2025)
- [Smooth Imitation Learning via Smooth Costs and Smooth Policies](https://mlsanity.com/paper/2111.02354/) (2021)
- [Inference for relative sparsity](https://mlsanity.com/paper/2306.14297/) (2023)

## Abstract (arXiv)

We study the problem of smooth imitation learning for online sequence prediction, where the goal is to train a policy that can smoothly imitate demonstrated behavior in a dynamic and continuous environment in response to online, sequential context input. Since the mapping from context to behavior is often complex, we take a learning reduction approach to reduce smooth imitation learning to a regression problem using complex function classes that are regularized to ensure smoothness. We present a learning meta-algorithm that achieves fast and stable convergence to a good policy. Our approach enjoys several attractive properties, including being fully deterministic, employing an adaptive learning rate that can provably yield larger policy improvements compared to previous approaches, and the ability to ensure stable convergence. Our empirical results demonstrate significant performance gains over previous approaches.