手法を比較

	自己教師あり強化学習	Semi-supervised Reinforcement Learning
分野	深層学習	深層学習
系統	Machine learning	Machine learning
提唱年	2020	2020s
提唱者	Laskin, M.; Srinivas, A.; Abbeel, P. (and contemporaries)	Multiple contributors (Laskin, Srinivas, Abbeel et al.)
種類	Self-supervised auxiliary-task learning for RL	Semi-supervised training paradigm for RL agents
原典	Laskin, M., Srinivas, A., & Abbeel, P. (2020). CURL: Contrastive Unsupervised Representations for Reinforcement Learning. Proceedings of the 37th International Conference on Machine Learning (ICML), PMLR 119, 5639–5650.	Zhan, X., Zhu, X., & Shi, H. (2022). Deepthermal: Combustion optimization for thermal power generating units using offline reinforcement learning. Proceedings of the AAAI Conference on Artificial Intelligence, 36(4), 4680–4688.
別名	SSL-RL, self-supervised RL, representation-based reinforcement learning, auxiliary-task RL	SSRL, semi-supervised RL, RL with unlabeled data, label-efficient reinforcement learning
関連	4	6
概要	Self-supervised Reinforcement Learning (SSL-RL) augments standard RL training with self-supervised auxiliary objectives — such as contrastive, predictive, or data-augmentation-based tasks — applied to the agent's own experience. These objectives improve the quality of learned representations without requiring extra human labels, enabling faster convergence and better sample efficiency, especially in high-dimensional observation spaces like raw pixels.	Semi-supervised reinforcement learning (SSRL) combines standard reinforcement learning — where an agent learns from sparse reward signals — with semi-supervised techniques that extract structure from unlabeled environment interactions. The goal is to improve sample efficiency and generalization when reward feedback is costly, delayed, or available only for a fraction of the agent's experience.