mlsanityData to 7 Oct 2026

MENTOR: Mixture-of-Experts Network with Task-Oriented Perturbation for Visual Reinforcement Learning

arXiv 2410.14972 ↗First posted 19 Oct 2024cs.RO cs.LGcited by 20

On the map: Mixture-of-experts → MoE in reinforcement learning and robotics

Abstract (arXiv)

Visual deep reinforcement learning (RL) enables robots to acquire skills from visual input for unstructured tasks. However, current algorithms suffer from low sample efficiency, limiting their practical applicability. In this work, we present MENTOR, a method that improves both the architecture and optimization of RL agents. Specifically, MENTOR replaces the standard multi-layer perceptron (MLP) with a mixture-of-experts (MoE) backbone and introduces a task-oriented perturbation mechanism. MENTOR outperforms state-of-the-art methods across three simulation benchmarks and achieves an average of 83% success rate on three challenging real-world robotic manipulation tasks, significantly surpassing the 32% success rate of the strongest existing model-free visual RL algorithm. These results underscore the importance of sample efficiency in advancing visual RL for real-world robotics. Experimental videos are available at https://suninghuang19.github.io/mentor_page/.

Key facts. Cited by 20 papers in the map (data through 7 Oct 2026). Placed in MoE in reinforcement learning and robotics, one of 12 directions in Mixture-of-experts. Builds on 37 of its 49 references that are in the map. First posted on arXiv 19 Oct 2024.

Builds on

37 of 49 references.

References · 37

  1. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,413
  2. 2021Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsitycited by 3,014
  3. 2020GShard: Scaling Giant Models with Conditional Computation and Automatic Shardingcited by 1,601
  4. 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,598
  5. 2020Gradient Surgery for Multi-Task Learningcited by 1,210
  6. 2023Mastering Diverse Domains through World Modelscited by 1,193
  7. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,168
  8. 2020Mastering Atari with Discrete World Modelscited by 990
  9. 2018DeepMind Control Suitecited by 935
  10. 2020CURL: Contrastive Unsupervised Representations for Reinforcement Learningcited by 745
  11. 2017Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrationscited by 685
  12. 2020Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixelscited by 573
25 more
  1. 2020Reinforcement Learning with Augmented Datacited by 452
  2. 2021Conflict-Averse Gradient Descent for Multi-task Learningcited by 441
  3. 2020Sim-to-Real Transfer in Deep Reinforcement Learning for Robotics: a Surveycited by 388
  4. 2022Temporal Difference Learning for Model Predictive Controlcited by 375
  5. 2021Mastering Visual Continuous Control: Improved Data-Augmented Reinforcement Learningcited by 359
  6. 2020Data-Efficient Reinforcement Learning with Self-Predictive Representationscited by 309
  7. 2017How to Escape Saddle Points Efficientlycited by 284
  8. 2019Challenges of Real-World Reinforcement Learningcited by 282
  9. 2020Decoupling Representation Learning from Reinforcement Learningcited by 263
  10. 2022The Primacy Bias in Deep Reinforcement Learningcited by 242
  11. 2023The Dormant Neuron Phenomenon in Deep Reinforcement Learningcited by 171
  12. 2018A Study on Overfitting in Deep Reinforcement Learningcited by 170
  13. 2015Adding Gradient Noise Improves Learning for Very Deep Networkscited by 144
  14. 2024SERL: A Software Suite for Sample-Efficient Robotic Reinforcement Learningcited by 113
  15. 2020The Ingredients of Real-World Robotic Reinforcement Learningcited by 111
  16. 2019Observational Overfitting in Reinforcement Learningcited by 90
  17. 2023TACO: Temporal Latent Action-Driven Contrastive Loss for Visual Reinforcement Learningcited by 56
  18. 2023DrM: Mastering Visual Reinforcement Learning through Dormant Ratio Minimizationcited by 45
  19. 2022Stabilizing Off-Policy Deep Reinforcement Learning from Pixelscited by 32
  20. 2023SpawnNet: Learning Generalizable Visuomotor Skills from Pre-trained Networkscited by 24
  21. 2022A Comprehensive Survey of Data Augmentation in Visual Reinforcement Learningcited by 23
  22. 2023Alternating Gradient Descent and Mixture-of-Experts for Integrated Multimodal Perceptioncited by 19
  23. 2024ACE : Off-Policy Actor-Critic with Causality-Aware Entropy Regularizationcited by 16
  24. 2023Improving Generalization in Visual Reinforcement Learning via Conflict-aware Gradient Agreement Augmentationcited by 13
  25. 2024Solving Token Gradient Conflict in Mixture-of-Experts for Large Vision-Language Modelcited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 10; 2026: 10.

20 papers.

Sort:

By area:

  1. 2025A Comprehensive Survey of Mixture-of-Experts: Algorithms, Theory, and Applicationscited by 91
  2. 2025DriveMoE: Mixture-of-Experts for Vision-Language-Action Model in End-to-End Autonomous Drivingcited by 85
  3. 2025TGRPO :Fine-tuning Vision-Language-Action Model via Trajectory-wise Group Relative Policy Optimizationcited by 47
  4. 2025MoE-Loco: Mixture of Experts for Multitask Locomotioncited by 17
  5. 2025Mastering Massive Multi-Task Reinforcement Learning via Mixture-of-Expert Decision Transformercited by 15
  6. 2026AtomicVLA: Unlocking the Potential of Atomic Skill Learning in Robotscited by 15
  7. 2025GRaD-Nav++: Vision-Language Model Enabled Visual Drone Navigation with Gaussian Radiance Fields and Differentiable Dynamicscited by 14
  8. 2025Expertise need not monopolize: Action-Specialized Mixture of Experts for Vision-Language-Action Learningcited by 12
  9. 2025MoE-DP: An MoE-Enhanced Diffusion Policy for Robust Long-Horizon Robotic Manipulation with Skill Decomposition and Failure Recoverycited by 9
  10. 2026Abstracting Robot Manipulation Skills via Mixture-of-Experts Diffusion Policiescited by 8
  11. 2025Mixture-of-Experts Meets In-Context Reinforcement Learningcited by 7
  12. 2026Breaking Lock-In: Preserving Steerability under Low-Data VLA Post-Trainingcited by 6
8 more
  1. 2026SPHERE: Mitigating the Loss of Spectral Plasticity in Mixture-of-Experts for Deep Reinforcement Learningcited by 2
  2. 2026DyGRO-VLA: Cross-Task Scaling of Vision-Language-Action Models via Dynamic Grouped Residual Optimizationcited by 2
  3. 2026SARM2: Multi-Task Stage Aware Reward Modeling for Self Improving Robotic Manipulationcited by 2
  4. 2026SkillMemo: Expert-guided Skill Memory Framework for Compositional Embodied Manipulationcited by 2
  5. 2025Dynamic Mixture of Experts Against Severe Distribution Shiftscited by 0
  6. 2026Robust Exploration in Directed Controller Synthesis via Reinforcement Learning with Soft Mixture-of-Expertscited by 0
  7. 2026Decoupling Exploration and Policy Optimization: Uncertainty Guided Tree Search for Hard Explorationcited by 0
  8. 2026V-Simba: Unleashing the Architectural Potential of RL in Visual Continuous Controlcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2022How to Spend Your Robot Time: Bridging Kickstarting and Offline Reinforcement Learning for Vision-based Robotic Manipulation
  2. 2025MoRE: Unlocking Scalability in Reinforcement Learning for Quadruped Vision-Language-Action Models
  3. 2022VRL3: A Data-Driven Framework for Visual Deep Reinforcement Learning
  4. 2026Z-1: Efficient Reinforcement Learning for Vision-Language-Action Models
  5. 2022Masked World Models for Visual Control
  6. 2021Robot Skill Adaptation via Soft Actor-Critic Gaussian Mixture Models
  7. 2022MoDem: Accelerating Visual Model-Based Reinforcement Learning with Demonstrations
  8. 2024Affordance-Guided Reinforcement Learning via Visual Prompting