paper-with-me

홈 › Papers

Task-Agnostic Continual Reinforcement Learning: Gaining Insights and Overcoming Challenges

2022-05-28 · Massimo Caccia, Jonas Mueller, Taesup Kim, Laurent Charlin, Rasool Fakoor

Continual learning (CL) enables the development of models and agents that learn from a sequence of tasks while addressing the limitations of standard deep learning approaches, such as catastrophic forgetting. In this work, we investigate the factors that contribute to the performance differences between task-agnostic CL and multi-task (MTL) agents. We pose two hypotheses: (1) task-agnostic methods might provide advantages in settings with limited data, computation, or high dimensionality, and (2) faster adaptation may be particularly beneficial in continual learning settings, helping to mitigate the effects of catastrophic forgetting. To investigate these hypotheses, we introduce a replay-based recurrent reinforcement learning (3RL) methodology for task-agnostic CL agents. We assess 3RL on a synthetic task and the Meta-World benchmark, which includes 50 unique manipulation tasks. Our results demonstrate that 3RL outperforms baseline methods and can even surpass its multi-task equivalent in challenging settings with high dimensionality. We also show that the recurrent task-agnostic agent consistently outperforms or matches the performance of its transformer-based counterpart. These findings provide insights into the advantages of task-agnostic CL over task-aware MTL approaches and highlight the potential of task-agnostic methods in resource-constrained, high-dimensional, and multi-task environments.

📄 PDF Abstract BibTeX arXiv:2205.14495

Code (2)

amazon-research/replay-based-recurrent-rl 공식 구현 pytorch
amazon-science/replay-based-recurrent-rl 공식 구현 pytorch

Tasks

Continual LearningContinuous ControlMulti-Task Learningreinforcement-learningReinforcement LearningReinforcement Learning (RL)

Methods 이 논문이 사용한 방법론

Experience Replay Experience Replay is a replay memory technique used in reinforcement learning where we store the agent’s experiences at each time-step, $e\_{t} = \left(s\_{t}, a\_{t}, r\_{t},…

Similar Papers 제목 키워드 기반

The Effectiveness of World Models for Continual Reinforcement Learning

2022-11-29 · Samuel Kessler, Mateusz Ostaszewski, Michał Bortkiewicz, Mateusz Żarski 외

World models power some of the most efficient reinforcement learning algorithms. In this work, we showcase that they can be harnessed for continual learning - a situation when the agent faces changing environments. World…

Continual LearningModel-based Reinforcement Learningreinforcement-learningReinforcement Learning+1

Knowledge Retention for Continual Model-Based Reinforcement Learning

2025-03-06 · Yixiang Sun, Haotian Fu, Michael Littman, George Konidaris

We propose DRAGO, a novel approach for continual model-based reinforcement learning aimed at improving the incremental development of world models across a sequence of tasks that differ in their reward functions but not …

Continual LearningModel-based Reinforcement Learningreinforcement-learningReinforcement Learning

On Diagnostics for Understanding Agent Training Behaviour in Cooperative MARL

2023-12-13 · Wiem Khlifi, Siddarth Singh, Omayma Mahjoub, Ruan de Kock 외

Cooperative multi-agent reinforcement learning (MARL) has made substantial strides in addressing the distributed decision-making challenges. However, as multi-agent systems grow in complexity, gaining a comprehensive und…

Decision MakingMulti-agent Reinforcement Learning

Reinforcement Learning from Diffusion Feedback: Q* for Image Search

2023-11-27 · Aboli Marathe

Large vision-language models are steadily gaining personalization capabilities at the cost of fine-tuning or data augmentation. We present two models for image generation using model-agnostic learning that align semantic…

Data AugmentationDiversityImage GenerationImage Retrieval+3

Advancements and Challenges in Continual Reinforcement Learning: A Comprehensive Review

2025-06-27 · Amara Zuffer, Michael Burke, Mehrtash Harandi

The diversity of tasks and dynamic nature of reinforcement learning (RL) require RL agents to be able to learn sequentially and continuously, a learning paradigm known as continuous reinforcement learning. This survey re…

Continual LearningDiversityreinforcement-learningReinforcement Learning+1