Shared-unique Features and Task-aware Prioritized Sampling on Multi-task Reinforcement Learning
Not on the map. Why?
Abstract (arXiv)
We observe that current state-of-the-art (SOTA) methods suffer from the performance imbalance issue when performing multi-task reinforcement learning (MTRL) tasks. While these methods may achieve impressive performance on average, they perform extremely poorly on a few tasks. To address this, we propose a new and effective method called STARS, which consists of two novel strategies: a shared-unique feature extractor and task-aware prioritized sampling. First, the shared-unique feature extractor learns both shared and task-specific features to enable better synergy of knowledge between different tasks. Second, the task-aware sampling strategy is combined with the prioritized experience replay for efficient learning on tasks with poor performance. The effectiveness and stability of our STARS are verified through experiments on the mainstream Meta-World benchmark. From the results, our STARS statistically outperforms current SOTA methods and alleviates the performance imbalance issue. Besides, we visualize the learned features to support our claims and enhance the interpretability of STARS.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 32 of its 45 references that are in the map. First posted on arXiv 2 Jun 2024.
Builds on
32 of 45 references.
References · 32
- 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,635
- 2015Continuous control with deep reinforcement learningcited by 3,279
- 2013Playing Atari with Deep Reinforcement Learningcited by 3,211
- 2015FaceNet: A Unified Embedding for Face Recognition and Clusteringcited by 2,833
- 2015Trust Region Policy Optimizationcited by 2,503
- 2015Deep Reinforcement Learning with Double Q-learningcited by 1,704
- 2020Gradient Surgery for Multi-Task Learningcited by 1,210
- 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,168
- 2017Hindsight Experience Replaycited by 1,096
- 2015Prioritized Experience Replaycited by 1,070
- 2017A Survey on Multi-Task Learningcited by 838
- 2018Multi-Task Learning as Multi-Objective Optimizationcited by 809
20 more
- 2017An Overview of Multi-Task Learning in Deep Neural Networkscited by 791
- 2018QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulationcited by 730
- 2020Deep Reinforcement Learning for Autonomous Driving: A Surveycited by 698
- 2016Deep Reinforcement Learning for Robotic Manipulation with Asynchronous Off-Policy Updatescited by 342
- 2016Modular Multitask Reinforcement Learning with Policy Sketchescited by 200
- 2016Sample Efficient Actor-Critic with Experience Replaycited by 185
- 2020Revisiting Fundamentals of Experience Replaycited by 171
- 2021Multi-Task Reinforcement Learning with Context-based Representationscited by 169
- 2020Multi-Task Reinforcement Learning with Soft Modularizationcited by 138
- 2016Learning Modular Neural Network Policies for Multi-Task and Multi-Robot Transfercited by 124
- 2023Synthetic Experience Replaycited by 100
- 2019Experience Replay Optimizationcited by 47
- 2024Sharing Knowledge in Multi-Task Deep Reinforcement Learningcited by 42
- 2022PaCo: Parameter-Compositional Multi-Task Reinforcement Learningcited by 40
- 2016Learning to Push by Grasping: Using multiple tasks for effective learningcited by 36
- 2022Multi-Task Imitation Learning for Linear Dynamical Systemscited by 22
- 2020Scalable Multi-Task Imitation Learning with Autonomous Improvementcited by 21
- 2016Learning Shared Representations in Multi-task Reinforcement Learningcited by 15
- 2022SKILL-IL: Disentangling Skill and Knowledge in Multitask Imitation Learningcited by 4
- 2021Evolutionary Multitask Optimization: Fundamental Research Questions, Practices, and Directions for the Futurecited by 3
Built on by
Not cited yet.