mlsanityData to 14 Oct 2026

MOCHA: Multi-Objective Co-Design using Hypernetwork Architectures

arXiv 2609.30570 ↗First posted 24 Sep 2026cs.ROcited by 0

On the map: Deep reinforcement learning methods → RL for games and combinatorial optimization

Abstract (arXiv)

In this work, we present MOCHA, the first, to our knowledge, reinforcement learning based approach to computing a family of Pareto-optimal policies across the design space of a robot using a single network. Specifically, MOCHA leverages the hypernetwork architecture to learn a network that produces specialized network parameters optimized for a given objective and parameterized robot design; we term this a multi-objective design hypernetwork (MDH). We demonstrate the capabilities of MDHs to represent a complex family of design-dependent strategies on two distinct robot morphologies, each with six design dimensions and across 2-3 objectives. Moreover, we propose an approach for efficiently producing a Design Pareto set using evolutionary search of the learned policy network, generating the optimal design-policy combination for each objective prioritization. Lastly, we provide an efficient method for computing generalist robot designs which achieve the best cumulative performance across the entire set of objectives.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for games and combinatorial optimization, one of 15 directions in Deep reinforcement learning methods. Builds on 27 of its 43 references that are in the map. First posted on arXiv 24 Sep 2026.

Builds on

27 of 43 references.

Closest ancestors · 2

Same direction, newest first

  1. 2026MO-Playground: Massively Parallelized Multi-Objective Reinforcement Learning for Robotics10 Mar
  2. 2025Pareto Set Learning for Multi-Objective Reinforcement Learning12 Jan

Other references · 25

  1. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,611
  2. 2015High-Dimensional Continuous Control Using Generalized Advantage Estimationcited by 2,026
  3. 2021Brax -- A Differentiable Physics Engine for Large Scale Rigid Body Simulationcited by 407
  4. 2020dm_control: Software and Tasks for Continuous Controlcited by 375
  5. 2019Scalable Global Optimization via Local Bayesian Optimizationcited by 374
  6. 2020pymoo: Multi-objective Optimization in Pythoncited by 200
  7. 2021Embodied Intelligence via Learning and Evolutioncited by 163
  8. 2017Preparing for the Unknown: Learning a Universal Policy with Online System Identificationcited by 141
  9. 2022PD-MORL: Preference-Driven Multi-Objective Reinforcement Learning Algorithmcited by 58
  10. 2022GenLoco: Generalized Locomotion Controllers for Quadrupedal Robotscited by 48
  11. 2022Evolution Gym: A Large-Scale Benchmark for Evolving Soft Robotscited by 47
  12. 2018Hardware Conditioned Policies for Multi-Robot Transfer Learningcited by 46
13 more
  1. 2024One Policy to Run Them All: an End-to-end Learning Approach to Multi-Embodiment Locomotioncited by 43
  2. 2022Hypernetworks in Meta-Reinforcement Learningcited by 42
  3. 2021Transform2Act: Learning a Transform-and-Control Policy for Efficient Agent Designcited by 38
  4. 2019Data-efficient Co-Adaptation of Morphology and Behaviour with Deep Reinforcement Learningcited by 37
  5. 2021Multi-Objective Graph Heuristic Search for Terrestrial Robot Designcited by 20
  6. 2022Hypernetworks for Zero-shot Transfer in Reinforcement Learningcited by 19
  7. 2023Recurrent Hypernetworks are Surprisingly Strong in Meta-RLcited by 16
  8. 2025AMOR: Adaptive Character Control through Multi-Objective Reinforcement Learningcited by 9
  9. 2022N-LIMB: Neural Limb Optimization for Efficient Morphological Designcited by 8
  10. 2024Learning Pareto Set for Multi-Objective Continuous Robot Controlcited by 8
  11. 2023Humanoid Robot Co-Design: Coupling Hardware Design with Gait Generation via Hybrid Zero Dynamicscited by 5
  12. 2025Toward Humanoid Brain-Body Co-design: Joint Optimization of Control and Morphology for Fall Recoverycited by 2
  13. 2026Shape Your Body: Value Gradients for Multi-Embodiment Robot Designcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2022C^2:Co-design of Robots via Concurrent Networks Coupling Online and Offline Reinforcement Learning
  2. 2025Capability-Aware Shared Hypernetworks for Flexible Heterogeneous Multi-Robot Coordination
  3. 2024A Talent-infused Policy-gradient Approach to Efficient Co-Design of Morphology and Task Allocation Behavior of Multi-Robot Systems ↗
  4. 2025Hypernetwork-based approach for optimal composition design in partially controlled multi-agent systems
  5. 2025Graph-Supported Dynamic Algorithm Configuration for Multi-Objective Combinatorial Optimization
  6. 2025Hyper-GoalNet: Goal-Conditioned Manipulation Policy Learning with HyperNetworks
  7. 2026Rapid co-design of Buoyancy-assisted robots for Challenging Locomotion using Gaussian Evolutionary Specialists ↗
  8. 2025Debate2Create: Robot Co-design via Multi-Agent LLM Debate