MODEL-BASED OFFLINE META-REINFORCEMENT LEARNING WITH REGULARIZATION

Social Sciences

• reinforcement 61%
• learning method 26%
• learning 21%
• quality good 10%
• performance 8%
• critic 8%
• experiment 5%
• evaluation 4%

Arts & Humanities

• Reinforcement Learning 100%
• Empirical Study 3%
• Intrinsic 3%
• Inference 3%
• Performance 2%
• Experiment 2%
• Evaluation 2%

Engineering & Materials Science

• Reinforcement learning 63%
• Experiments 3%