手法を比較

	ドメイン適応型強化学習	深層強化学習	転移学習
分野	深層学習	深層学習	機械学習
系統	Machine learning	Machine learning	Machine learning
提唱年	2009–2020	2015	2010 (formalized); 1990s (early roots)
提唱者	Multiple contributors (Taylor & Stone 2009 survey; Kim et al. 2020 among key formalizations)	Mnih, V. et al. (DQN)	Pan, S. J. & Yang, Q. (survey); Bengio, Y. (deep learning framing)
種類	Transfer-based RL paradigm	Sequential decision-making (agent–environment interaction)	Learning paradigm
原典	Kim, K., Kim, H., Lim, H., & Choi, J. (2020). Domain Adaptive Reinforcement Learning with Model-Based Approach. arXiv preprint arXiv:2102.03170.	Mnih, V. et al. (2015). Human-Level Control through Deep Reinforcement Learning. Nature, 518, 529–533.	Pan, S. J., & Yang, Q. (2010). A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering, 22(10), 1345–1359.
別名	Domain-Adaptive RL, DARL, Cross-domain RL, Transfer RL with domain adaptation	Derin Pekiştirmeli Öğrenme (DQN / PPO / A3C), derin pekiştirmeli öğrenme, deep RL, DRL	TL, domain adaptation, fine-tuning, pre-trained model adaptation
関連	2	4	3
概要	Domain-Adaptive Reinforcement Learning (DARL) extends standard RL by enabling a policy trained in one environment or domain to transfer and generalise effectively to a different but related target domain. It addresses the domain-shift problem — where dynamics, observations, or reward structures differ between training and deployment — through alignment, adaptation, or domain-randomisation techniques, reducing the need to collect costly experience in the target domain.	Deep Reinforcement Learning combines neural networks with reinforcement learning so an agent learns by interacting with an environment, popularised by Mnih and colleagues' 2015 Nature work on human-level Atari control. Instead of learning from a fixed labelled dataset, the agent takes actions, observes rewards, and gradually shapes a policy that maximises long-run return.	Transfer learning is a machine learning paradigm in which knowledge gained from training a model on a source task or domain is reused to improve learning on a different but related target task or domain. It is especially powerful when labeled data for the target task is scarce, and it underlies most modern deep learning applications in computer vision, natural language processing, and beyond.