# Accelerating Transformers in Online RL

arXiv: https://arxiv.org/abs/2509.26137 · HTML: https://mlsanity.com/paper/2509.26137/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 34 of its 42 references that are in the map.
- First posted on arXiv 30 Sep 2025.

## Placed in

- Deep reinforcement learning methods / [RL for robot locomotion and manipulation](https://mlsanity.com/deeprl/rl-for-robot-locomotion-and-manipulation/)

## Builds on (most cited first, 30 of 34)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Continuous control with deep reinforcement learning](https://mlsanity.com/paper/1509.02971/) (2015)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [Addressing Function Approximation Error in Actor-Critic Methods](https://mlsanity.com/paper/1802.09477/) (2018)
- [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context](https://mlsanity.com/paper/1901.02860/) (2019)
- [Decision Transformer: Reinforcement Learning via Sequence Modeling](https://mlsanity.com/paper/2106.01345/) (2021)
- [Octo: An Open-Source Generalist Robot Policy](https://mlsanity.com/paper/2405.12213/) (2024)
- [A Generalist Agent](https://mlsanity.com/paper/2205.06175/) (2022)
- [AWAC: Accelerating Online Reinforcement Learning with Offline Datasets](https://mlsanity.com/paper/2006.09359/) (2020)
- [Offline Reinforcement Learning as One Big Sequence Modeling Problem](https://mlsanity.com/paper/2106.02039/) (2021)
- [VIMA: General Robot Manipulation with Multimodal Prompts](https://mlsanity.com/paper/2210.03094/) (2022)
- [Stabilizing Transformers for Reinforcement Learning](https://mlsanity.com/paper/1910.06764/) (2019)
- [ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI](https://mlsanity.com/paper/2410.00425/) (2024)
- [Multi-Game Decision Transformers](https://mlsanity.com/paper/2205.15241/) (2022)
- [Recurrent Memory Transformer](https://mlsanity.com/paper/2207.06881/) (2022)
- [Q-Transformer: Scalable Offline Reinforcement Learning via Autoregressive Q-Functions](https://mlsanity.com/paper/2309.10150/) (2023)
- [Q-learning Decision Transformer: Leveraging Dynamic Programming for Conditional Sequence Modelling in Offline RL](https://mlsanity.com/paper/2209.03993/) (2022)
- [Policy Agnostic RL: Offline RL and Online RL Fine-Tuning of Any Class and Backbone](https://mlsanity.com/paper/2412.06685/) (2024)
- [Retrieval-Augmented Reinforcement Learning](https://mlsanity.com/paper/2202.08417/) (2022)
- [SMART: Self-supervised Multi-task pretrAining with contRol Transformers](https://mlsanity.com/paper/2301.09816/) (2023)
- [STRAP: Robot Sub-Trajectory Retrieval for Augmented Policy Learning](https://mlsanity.com/paper/2412.15182/) (2024)
- [Bootstrapped Transformer for Offline Reinforcement Learning](https://mlsanity.com/paper/2206.08569/) (2022)
- [Deep Transformer Q-Networks for Partially Observable Reinforcement Learning](https://mlsanity.com/paper/2206.01078/) (2022)
- [Efficient Transformers in Reinforcement Learning using Actor-Learner Distillation](https://mlsanity.com/paper/2104.01655/) (2021)
- [TOP-ERL: Transformer-based Off-Policy Episodic Reinforcement Learning](https://mlsanity.com/paper/2410.09536/) (2024)
- [Recurrent Action Transformer with Memory](https://mlsanity.com/paper/2306.09459/) (2023)
- [Efficient Recurrent Off-Policy RL Requires a Context-Encoder-Specific Learning Rate](https://mlsanity.com/paper/2405.15384/) (2024)
- [QT-TDM: Planning With Transformer Dynamics Model and Autoregressive Q-Learning](https://mlsanity.com/paper/2407.18841/) (2024)

## Abstract (arXiv)

The appearance of transformer-based models in Reinforcement Learning (RL) has expanded the horizons of possibilities in robotics tasks, but it has simultaneously brought a wide range of challenges during its implementation, especially in model-free online RL. Some of the existing learning algorithms cannot be easily implemented with transformer-based models due to the instability of the latter. In this paper, we propose a method that uses the Accelerator policy as a transformer's trainer. The Accelerator, a simpler and more stable model, interacts with the environment independently while simultaneously training the transformer through behavior cloning during the first stage of the proposed algorithm. In the second stage, the pretrained transformer starts to interact with the environment in a fully online setting. As a result, this model-free algorithm accelerates the transformer in terms of its performance and helps it to train online in a more stable and faster way. By conducting experiments on both state-based and image-based ManiSkill environments, as well as on MuJoCo tasks in MDP and POMDP settings, we show that applying our algorithm not only enables stable training of transformers but also reduces training time on image-based environments by up to a factor of two. Moreover, it decreases the required replay buffer size in off-policy methods to 10-20 thousand, which significantly lowers the overall computational demands.