# Learning from Symmetry: Meta-Reinforcement Learning with Symmetrical Behaviors and Language Instructions

arXiv: https://arxiv.org/abs/2209.10656 · HTML: https://mlsanity.com/paper/2209.10656/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 29 references that are in the map.
- First posted on arXiv 21 Sep 2022.

## Builds on (most cited first, 14 of 14)

- [Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks](https://mlsanity.com/paper/1703.03400/) (2017)
- [Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning](https://mlsanity.com/paper/1910.10897/) (2019)
- [CLIPort: What and Where Pathways for Robotic Manipulation](https://mlsanity.com/paper/2109.12098/) (2021)
- [BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learning](https://mlsanity.com/paper/2202.02005/) (2022)
- [Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variables](https://mlsanity.com/paper/1903.08254/) (2019)
- [MDP Homomorphic Networks: Group Symmetries in Reinforcement Learning](https://mlsanity.com/paper/2006.16908/) (2020)
- [Meta-Learning Symmetries by Reparameterization](https://mlsanity.com/paper/2007.02933/) (2020)
- [Guiding Policies with Language via Meta-Learning](https://mlsanity.com/paper/1811.07882/) (2018)
- [Extending a Parser to Distant Domains Using a Few Dozen Partially Annotated Examples](https://mlsanity.com/paper/1805.06556/) (2018)
- [PixL2R: Guiding Reinforcement Learning Using Natural Language by Mapping Pixels to Rewards](https://mlsanity.com/paper/2007.15543/) (2020)
- [Watch, Try, Learn: Meta-Learning from Demonstrations and Reward](https://mlsanity.com/paper/1906.03352/) (2019)
- [Introducing Symmetries to Black Box Meta Reinforcement Learning](https://mlsanity.com/paper/2109.10781/) (2021)
- [Meta-Reinforcement Learning via Language Instructions](https://mlsanity.com/paper/2209.04924/) (2022)
- [Complex Robotic Manipulation via Graph-Based Hindsight Goal Generation](https://mlsanity.com/paper/2007.13486/) (2020)

## Built on by (most cited first, 8 of 8)

- [Language-Conditioned Imitation Learning with Base Skill Priors under Unstructured Data](https://mlsanity.com/paper/2305.19075/) (2023)
- [Pick-and-place Manipulation Across Grippers Without Retraining: A Learning-optimization Diffusion Policy Approach](https://mlsanity.com/paper/2502.15613/) (2025)
- [Dreaming the Unseen: World Model-regularized Diffusion Policy for Out-of-Distribution Robustness](https://mlsanity.com/paper/2603.21017/) (2026)
- [Contact Energy Based Hindsight Experience Prioritization](https://mlsanity.com/paper/2312.02677/) (2023)
- [Inference-stage Adaptation-projection Strategy Adapts Diffusion Policy to Cross-manipulators Scenarios](https://mlsanity.com/paper/2509.11621/) (2025)
- [Real-Time Adaptive Safety-Critical Control with Gaussian Processes in High-Order Uncertain Models](https://mlsanity.com/paper/2402.18946/) (2024)
- [Time Reversal Symmetry for Efficient Robotic Manipulations in Deep Reinforcement Learning](https://mlsanity.com/paper/2505.13925/) (2025)
- [From Trajectories to Instructions: Language-Conditioned Meta-Reinforcement Learning](https://mlsanity.com/paper/2607.18830/) (2026)

## Abstract (arXiv)

Meta-reinforcement learning (meta-RL) is a promising approach that enables the agent to learn new tasks quickly. However, most meta-RL algorithms show poor generalization in multi-task scenarios due to the insufficient task information provided only by rewards. Language-conditioned meta-RL improves the generalization capability by matching language instructions with the agent's behaviors. While both behaviors and language instructions have symmetry, which can speed up human learning of new knowledge. Thus, combining symmetry and language instructions into meta-RL can help improve the algorithm's generalization and learning efficiency. We propose a dual-MDP meta-reinforcement learning method that enables learning new tasks efficiently with symmetrical behaviors and language instructions. We evaluate our method in multiple challenging manipulation tasks, and experimental results show that our method can greatly improve the generalization and learning efficiency of meta-reinforcement learning. Videos are available at https://tumi6robot.wixsite.com/symmetry/.