# Disentangled Skill Embeddings for Reinforcement Learning

arXiv: https://arxiv.org/abs/1906.09223 · HTML: https://mlsanity.com/paper/1906.09223/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Jun 2019.

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

- [Unsupervised Domain Adaptation with Dynamics-Aware Rewards in Reinforcement Learning](https://mlsanity.com/paper/2110.12997/) (2021)
- [Multi-Agent Policy Transfer via Task Relationship Modeling](https://mlsanity.com/paper/2203.04482/) (2022)
- [Uncertainty Estimation Using Riemannian Model Dynamics for Offline Reinforcement Learning](https://mlsanity.com/paper/2102.11327/) (2021)
- [On the relationship between disentanglement and multi-task learning](https://mlsanity.com/paper/2110.03498/) (2021)
- [A Survey on Deep Reinforcement Learning-based Approaches for Adaptation and Generalization](https://mlsanity.com/paper/2202.08444/) (2022)
- [Variational Inference for Model-Free and Model-Based Reinforcement Learning](https://mlsanity.com/paper/2209.01693/) (2022)
- [Learning to Transfer: A Foliated Theory](https://mlsanity.com/paper/2107.10763/) (2021)
- [Transfer Reinforcement Learning for Differing Action Spaces via Q-Network Representations](https://mlsanity.com/paper/2202.02442/) (2022)
- [Disentangled Skill Representations for Predictive Human Modeling](https://mlsanity.com/paper/2608.23776/) (2026)

## Abstract (arXiv)

We propose a novel framework for multi-task reinforcement learning (MTRL). Using a variational inference formulation, we learn policies that generalize across both changing dynamics and goals. The resulting policies are parametrized by shared parameters that allow for transfer between different dynamics and goal conditions, and by task-specific latent-space embeddings that allow for specialization to particular tasks. We show how the latent-spaces enable generalization to unseen dynamics and goals conditions. Additionally, policies equipped with such embeddings serve as a space of skills (or options) for hierarchical reinforcement learning. Since we can change task dynamics and goals independently, we name our framework Disentangled Skill Embeddings (DSE).