# Learning Goal Embeddings via Self-Play for Hierarchical Reinforcement Learning

arXiv: https://arxiv.org/abs/1811.09083 · HTML: https://mlsanity.com/paper/1811.09083/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Nov 2018.

## Built on by (most cited first, 24 of 24)

- [Asymmetric self-play for automatic goal discovery in robotic manipulation](https://mlsanity.com/paper/2101.04882/) (2021)
- [Landmark-Guided Subgoal Generation in Hierarchical Reinforcement Learning](https://mlsanity.com/paper/2110.13625/) (2021)
- [Contextualize Me -- The Case for Context in Reinforcement Learning](https://mlsanity.com/paper/2202.04500/) (2022)
- [Multi-task curriculum learning in a complex, visual, hard-exploration domain: Minecraft](https://mlsanity.com/paper/2106.14876/) (2021)
- [Learning Goal-Conditioned Policies Offline with Self-Supervised Reward Shaping](https://mlsanity.com/paper/2301.02099/) (2023)
- [Learn Goal-Conditioned Policy with Intrinsic Motivation for Deep Reinforcement Learning](https://mlsanity.com/paper/2104.05043/) (2021)
- [Understanding the Complexity Gains of Single-Task RL with a Curriculum](https://mlsanity.com/paper/2212.12809/) (2022)
- [RL for Reasoning by Adaptively Revealing Rationales](https://mlsanity.com/paper/2506.18110/) (2025)
- [It Takes Four to Tango: Multiagent Selfplay for Automatic Curriculum Generation](https://mlsanity.com/paper/2202.10608/) (2022)
- [Discrete Factorial Representations as an Abstraction for Goal Conditioned Reinforcement Learning](https://mlsanity.com/paper/2211.00247/) (2022)
- [Learning to Drive via Asymmetric Self-Play](https://mlsanity.com/paper/2409.18218/) (2024)
- [Probabilistic Subgoal Representations for Hierarchical Reinforcement learning](https://mlsanity.com/paper/2406.16707/) (2024)
- [Augmenting GAIL with BC for sample efficient imitation learning](https://mlsanity.com/paper/2001.07798/) (2020)
- [Towards Skilled Population Curriculum for Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2302.03429/) (2023)
- [Walk the Random Walk: Learning to Discover and Reach Goals Without Supervision](https://mlsanity.com/paper/2206.11733/) (2022)
- [Active Task Randomization: Learning Robust Skills via Unsupervised Generation of Diverse and Feasible Tasks](https://mlsanity.com/paper/2211.06134/) (2022)
- [Unsupervised Skill Discovery for Robotic Manipulation through Automatic Task Generation](https://mlsanity.com/paper/2410.04855/) (2024)
- [Eden: A Unified Environment Framework for Booming Reinforcement Learning Algorithms](https://mlsanity.com/paper/2109.01768/) (2021)
- [A framework for the emergence and analysis of language in social learning agents](https://mlsanity.com/paper/2305.02632/) (2023)
- [Scalable Interactive Machine Learning for Future Command and Control](https://mlsanity.com/paper/2402.06501/) (2024)
- [CORD: Generalizable Cooperation via Role Diversity](https://mlsanity.com/paper/2501.02221/) (2025)
- [AutoDIME: Automatic Design of Interesting Multi-Agent Environments](https://mlsanity.com/paper/2203.02481/) (2022)
- [Unsupervised Learning of Efficient Exploration: Pre-training Adaptive Policies via Self-Imposed Goals](https://mlsanity.com/paper/2601.19810/) (2026)
- [Learning to Play: Reinforcement Learning and Games](https://mlsanity.com/paper/2610.17391/) (2026)

## Abstract (arXiv)

In hierarchical reinforcement learning a major challenge is determining appropriate low-level policies. We propose an unsupervised learning scheme, based on asymmetric self-play from Sukhbaatar et al. (2018), that automatically learns a good representation of sub-goals in the environment and a low-level policy that can execute them. A high-level policy can then direct the lower one by generating a sequence of continuous sub-goal vectors. We evaluate our model using Mazebase and Mujoco environments, including the challenging AntGather task. Visualizations of the sub-goal embeddings reveal a logical decomposition of tasks within the environment. Quantitatively, our approach obtains compelling performance gains over non-hierarchical approaches.