# Composing Diffusion Policies for Few-shot Learning of Movement Trajectories

arXiv: https://arxiv.org/abs/2410.17479 · HTML: https://mlsanity.com/paper/2410.17479/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models.
- Builds on 17 of its 28 references that are in the map.
- First posted on arXiv 22 Oct 2024.

## Placed in

- Diffusion models / [Diffusion and flow policies for robotics](https://mlsanity.com/diffusion/diffusion-and-flow-policies-for-robotics/)

## Builds on (most cited first, 17 of 17)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Deep Unsupervised Learning using Nonequilibrium Thermodynamics](https://mlsanity.com/paper/1503.03585/) (2015)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [The Option-Critic Architecture](https://mlsanity.com/paper/1609.05140/) (2016)
- [Understanding Diffusion Models: A Unified Perspective](https://mlsanity.com/paper/2208.11970/) (2022)
- [Reduce, Reuse, Recycle: Compositional Generation with Energy-Based Diffusion Models and MCMC](https://mlsanity.com/paper/2302.11552/) (2023)
- [Learning Fast Samplers for Diffusion Models by Differentiating Through Sample Quality](https://mlsanity.com/paper/2202.05830/) (2022)
- [MCP: Learning Composable Hierarchical Control with Multiplicative Compositional Policies](https://mlsanity.com/paper/1905.09808/) (2019)
- [Measuring Compositionality in Representation Learning](https://mlsanity.com/paper/1902.07181/) (2019)
- [Learning to Compose Visual Relations](https://mlsanity.com/paper/2111.09297/) (2021)
- [LOTUS: Continual Imitation Learning for Robot Manipulation Through Unsupervised Skill Discovery](https://mlsanity.com/paper/2311.02058/) (2023)
- [PoCo: Policy Composition from and for Heterogeneous Robot Learning](https://mlsanity.com/paper/2402.02511/) (2024)
- [Compositional Diffusion-Based Continuous Constraint Solvers](https://mlsanity.com/paper/2309.00966/) (2023)
- [How to Reuse and Compose Knowledge for a Lifetime of Tasks: A Survey on Continual Learning and Functional Composition](https://mlsanity.com/paper/2207.07730/) (2022)
- [Composable Part-Based Manipulation](https://mlsanity.com/paper/2405.05876/) (2024)
- [Forward Kinematics Kernel for Improved Proxy Collision Checking](https://mlsanity.com/paper/1910.06451/) (2019)

## Built on by (most cited first, 3 of 3)

- [Generative Trajectory Stitching through Diffusion Composition](https://mlsanity.com/paper/2503.05153/) (2025)
- [Factored Diffusion Policies:Compositionally Generalized Robot Control with a Single Score Network](https://mlsanity.com/paper/2605.22596/) (2026)
- [MSG: Multi-Stream Generative Policies for Sample-Efficient Robotic Manipulation](https://mlsanity.com/paper/2509.24956/) (2025)

## Abstract (arXiv)

Humans can perform various combinations of physical skills without having to relearn skills from scratch every single time. For example, we can swing a bat when walking without having to re-learn such a policy from scratch by composing the individual skills of walking and bat swinging. Enabling robots to combine or compose skills is essential so they can learn novel skills and tasks faster with fewer real world samples. To this end, we propose a novel compositional approach called DSE- Diffusion Score Equilibrium that enables few-shot learning for novel skills by utilizing a combination of base policy priors. Our method is based on probabilistically composing diffusion policies to better model the few-shot demonstration data-distribution than any individual policy. Our goal here is to learn robot motions few-shot and not necessarily goal oriented trajectories. Unfortunately we lack a general purpose metric to evaluate the error between a skill or motion and the provided demonstrations. Hence, we propose a probabilistic measure - Maximum Mean Discrepancy on the Forward Kinematics Kernel (MMD-FK), that is task and action space agnostic. By using our few-shot learning approach DSE, we show that we are able to achieve a reduction of over 30% in MMD-FK across skills and number of demonstrations. Moreover, we show the utility of our approach through real world experiments by teaching novel trajectories to a robot in 5 demonstrations.