# Decoupling Dynamics and Reward for Transfer Learning

arXiv: https://arxiv.org/abs/1804.10689 · HTML: https://mlsanity.com/paper/1804.10689/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Apr 2018.

## Built on by (most cited first, 23 of 23)

- [For SALE: State-Action Representation Learning for Deep Reinforcement Learning](https://mlsanity.com/paper/2306.02451/) (2023)
- [Learning Markov State Abstractions for Deep Reinforcement Learning](https://mlsanity.com/paper/2106.04379/) (2021)
- [Towards General-Purpose Model-Free Reinforcement Learning](https://mlsanity.com/paper/2501.16142/) (2025)
- [Which Mutual-Information Representation Learning Objectives are Sufficient for Control?](https://mlsanity.com/paper/2106.07278/) (2021)
- [INFOrmation Prioritization through EmPOWERment in Visual Model-Based RL](https://mlsanity.com/paper/2204.08585/) (2022)
- [Towards Better Laplacian Representation in Reinforcement Learning with Generalized Graph Drawing](https://mlsanity.com/paper/2107.05545/) (2021)
- [Simplified Temporal Consistency Reinforcement Learning](https://mlsanity.com/paper/2306.09466/) (2023)
- [Multi-Agent Policy Transfer via Task Relationship Modeling](https://mlsanity.com/paper/2203.04482/) (2022)
- [Reachability-Aware Laplacian Representation in Reinforcement Learning](https://mlsanity.com/paper/2210.13153/) (2022)
- [A Unified View on Solving Objective Mismatch in Model-Based Reinforcement Learning](https://mlsanity.com/paper/2310.06253/) (2023)
- [Fractional Transfer Learning for Deep Model-Based Reinforcement Learning](https://mlsanity.com/paper/2108.06526/) (2021)
- [Lifelong Hyper-Policy Optimization with Multiple Importance Sampling Regularization](https://mlsanity.com/paper/2112.06625/) (2021)
- [Multi-Source Transfer Learning for Deep Model-Based Reinforcement Learning](https://mlsanity.com/paper/2205.14410/) (2022)
- [Task and Domain Adaptive Reinforcement Learning for Robot Control](https://mlsanity.com/paper/2404.18713/) (2024)
- [Debiased Model-based Representations for Sample-efficient Continuous Control](https://mlsanity.com/paper/2605.11711/) (2026)
- [Transfer Reinforcement Learning for Differing Action Spaces via Q-Network Representations](https://mlsanity.com/paper/2202.02442/) (2022)
- [Reinforcement Learning in the Wild with Maximum Likelihood-based Model Transfer](https://mlsanity.com/paper/2302.09273/) (2023)
- [Overcoming Slow Decision Frequencies in Continuous Control: Model-Based Sequence Reinforcement Learning for Model-Free Control](https://mlsanity.com/paper/2410.08979/) (2024)
- [A Pre-Trained Graph-Based Model for Adaptive Sequencing of Educational Documents](https://mlsanity.com/paper/2411.11520/) (2024)
- [Goals and the Structure of Experience](https://mlsanity.com/paper/2508.15013/) (2025)
- [Sensitivity Shaping for Latent Modeling](https://mlsanity.com/paper/2606.14585/) (2026)
- [NASDAQ: Normalized Observation Space Dynamics-Augmented Q-Learning](https://mlsanity.com/paper/2606.21297/) (2026)
- [Reward as Observation: Learning Reward-Based Policies for Rapid Adaptation](https://mlsanity.com/paper/2610.00729/) (2026)

## Abstract (arXiv)

Current reinforcement learning (RL) methods can successfully learn single tasks but often generalize poorly to modest perturbations in task domain or training procedure. In this work, we present a decoupled learning strategy for RL that creates a shared representation space where knowledge can be robustly transferred. We separate learning the task representation, the forward dynamics, the inverse dynamics and the reward function of the domain, and show that this decoupling improves performance within the task, transfers well to changes in dynamics and reward, and can be effectively used for online planning. Empirical results show good performance in both continuous and discrete RL domains.