# Jointly Learning to Construct and Control Agents using Deep Reinforcement Learning

arXiv: https://arxiv.org/abs/1801.01432 · HTML: https://mlsanity.com/paper/1801.01432/

## Key facts

- Cited by 55 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Jan 2018.

## Built on by (most cited first, 30 of 55)

- [Embodied Intelligence via Learning and Evolution](https://mlsanity.com/paper/2102.02202/) (2021)
- [Scaling Cross-Embodied Learning: One Policy for Manipulation, Navigation, Locomotion and Aviation](https://mlsanity.com/paper/2408.11812/) (2024)
- [MetaMorph: Learning Universal Controllers with Transformers](https://mlsanity.com/paper/2203.11931/) (2022)
- [An End-to-End Differentiable Framework for Contact-Aware Robot Design](https://mlsanity.com/paper/2107.07501/) (2021)
- [Evolution Gym: A Large-Scale Benchmark for Evolving Soft Robots](https://mlsanity.com/paper/2201.09863/) (2022)
- [Transform2Act: Learning a Transform-and-Control Policy for Efficient Agent Design](https://mlsanity.com/paper/2110.03659/) (2021)
- [Task-Agnostic Morphology Evolution](https://mlsanity.com/paper/2102.13100/) (2021)
- [Universal Morphology Control via Contextual Modulation](https://mlsanity.com/paper/2302.11070/) (2023)
- [Bayesian Meta-Learning for Few-Shot Policy Adaptation Across Robotic Platforms](https://mlsanity.com/paper/2103.03697/) (2021)
- [Learning Modular Robot Control Policies](https://mlsanity.com/paper/2105.10049/) (2021)
- [Searching for High-Value Molecules Using Reinforcement Learning and Transformers](https://mlsanity.com/paper/2310.02902/) (2023)
- [REvolveR: Continuous Evolutionary Models for Robot-to-robot Policy Transfer](https://mlsanity.com/paper/2202.05244/) (2022)
- [From Universal Humanoid Control to Automatic Physically Valid Character Creation](https://mlsanity.com/paper/2206.09286/) (2022)
- [The Effects of Learning in Morphologically Evolving Robot Systems](https://mlsanity.com/paper/2111.09851/) (2021)
- [VLMgineer: Vision Language Models as Robotic Toolsmiths](https://mlsanity.com/paper/2507.12644/) (2025)
- [DiffAqua: A Differentiable Computational Design Pipeline for Soft Underwater Swimmers with Shape Interpolation](https://mlsanity.com/paper/2104.00837/) (2021)
- [Dynamics-Guided Diffusion Model for Sensor-less Robot Manipulator Design](https://mlsanity.com/paper/2402.15038/) (2024)
- [MORPH: Design Co-optimization with Reinforcement Learning via a Differentiable Hardware Model Proxy](https://mlsanity.com/paper/2309.17227/) (2023)
- [Reinforcement learning for freeform robot design](https://mlsanity.com/paper/2310.05670/) (2023)
- [Text2Robot: Evolutionary Robot Design from Text Descriptions](https://mlsanity.com/paper/2406.19963/) (2024)
- [Learning to Design and Use Tools for Robotic Manipulation](https://mlsanity.com/paper/2311.00754/) (2023)
- [Efficient Hyperparameter Optimization for Physics-based Character Animation](https://mlsanity.com/paper/2104.12365/) (2021)
- [Leveraging Hyperbolic Embeddings for Coarse-to-Fine Robot Design](https://mlsanity.com/paper/2311.00462/) (2023)
- [Co-Optimizing Reconfigurable Environments and Policies for Decentralized Multi-Agent Navigation](https://mlsanity.com/paper/2403.14583/) (2024)
- [House of Dextra: Cross-embodied Co-design for Dexterous Hands](https://mlsanity.com/paper/2512.03743/) (2025)
- [Creative Problem Solving in Artificially Intelligent Agents: A Survey and Framework](https://mlsanity.com/paper/2204.10358/) (2022)
- [Task2Morph: Differentiable Task-inspired Framework for Contact-Aware Robot Design](https://mlsanity.com/paper/2403.19093/) (2024)
- [Embodied Co-Design for Rapidly Evolving Agents: Taxonomy, Frontiers, and Challenges](https://mlsanity.com/paper/2512.04770/) (2025)
- [Co-Imitation: Learning Design and Behaviour by Imitation](https://mlsanity.com/paper/2209.01207/) (2022)
- [Reinforcement Learning for Efficient Design and Control Co-optimisation of Energy Systems](https://mlsanity.com/paper/2406.19825/) (2024)

## Abstract (arXiv)

The physical design of a robot and the policy that controls its motion are inherently coupled, and should be determined according to the task and environment. In an increasing number of applications, data-driven and learning-based approaches, such as deep reinforcement learning, have proven effective at designing control policies. For most tasks, the only way to evaluate a physical design with respect to such control policies is empirical--i.e., by picking a design and training a control policy for it. Since training these policies is time-consuming, it is computationally infeasible to train separate policies for all possible designs as a means to identify the best one. In this work, we address this limitation by introducing a method that performs simultaneous joint optimization of the physical design and control network. Our approach maintains a distribution over designs and uses reinforcement learning to optimize a control policy to maximize expected reward over the design distribution. We give the controller access to design parameters to allow it to tailor its policy to each design in the distribution. Throughout training, we shift the distribution towards higher-performing designs, eventually converging to a design and control policy that are jointly optimal. We evaluate our approach in the context of legged locomotion, and demonstrate that it discovers novel designs and walking gaits, outperforming baselines in both performance and efficiency.