Transfer Learning in Deep Reinforcement Learning: A Survey

By Zhuangdi Zhu, Kaixiang Lin, Anil K. Jain, Jiayu Zhou. 2020

899 citations recorded by the source corpus.

Reinforcement learning is a learning paradigm for solving sequential decision-making problems. Recent years have witnessed remarkable progress in reinforcement learning upon the fast development of deep neural networks. Along with the promising prospects of reinforcement learning in numerous domains such as robotics and game-playing, transfer learning has arisen to tackle various challenges faced by reinforcement learning, by transferring knowledge from external expertise to facilitate the efficiency and effectiveness of the learning process. In this survey, we systematically investigate the recent progress of transfer learning approaches in the context of deep reinforcement learning. Specifically, we provide a framework for categorizing the state-of-the-art transfer learning approaches, under which we analyze their goals, methodologies, compatible reinforcement learning backbones, and practical applications. We also draw connections between transfer learning and other relevant topics from the reinforcement learning perspective and explore their potential challenges that await future research progress.

Subfields: Artificial Intelligence, Computer Science Applications

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Science Applications

Citations in the corpus

Disentangling Transfer in Continual Reinforcement Learning (2022, methodology): “[42, 51, 43, 40] present methods on how to find and use mappings between different domains.”
Block Contextual MDPs for Continual Learning (2021, methodology): “Meld Codebase: https://github.com/tonyzhaozh/meld 5.”
Structural Similarity for Improved Transfer in Reinforcement Learning (2022, neutral): “One potential solution to both of these problems is to leverage transfer learning [5, 15, 29], i.e., using previously trained policies to jumpstart learning in a new task, or by developing curricula [3, 13] of tasks to facilitate the learning of advanced skills or behaviors.”
A Review of Deep Reinforcement Learning for Smart Building Energy Management (2020, methodology): “To improve this situation, a possible way is to combine DRL with transfer learning [127].”
The rise and potential of large language model based agents: a survey (2025, positive): “Consequently, researchers introduce transfer learning to expedite an agent’s learning on new tasks [57–59].”
AdaRL: What, Where, and How to Adapt in Transfer Reinforcement Learning (2021, background): “There are roughly two research lines in transfer RL (Taylor and Stone, 2009; Zhu et al., 2020): (1) ﬁnding policies that are robust to environment variations, and (2) adapting policies from the source domain to the target domain as efﬁciently as possible.”
Reinforcement Learning Control of a Forestry Crane Manipulator (2021, background): “The concept of transfer learning (TL) [14] has shown potential to solve high sample complexity issues symptomatic”