Reinforcement Learning in Low-rank MDPs with Density Features

Keyphrases

• Reinforcement Learning 100%
• Density Feature 100%
• Low-rank MDPs 100%
• Learning Value 33%
• Occupancy 33%
• Unsupervised Learning 33%
• Representative Structures 33%
• Data Distribution 33%
• Low-rank 33%
• Learning Settings 33%
• Blowing 33%
• Function Approximation 33%
• Reachability 33%
• Online Algorithms 33%
• Technical Tools 33%
• Transition Matrix 33%
• Additive Error 33%
• Data Coverage 33%
• Representation Learning 33%
• Occupancy Estimation 33%
• Expressive Function 33%
• Sample-efficient Learning 33%

Computer Science

• Representation Learning 100%
• Unsupervised Learning 100%
• Data Distribution 100%
• Technical Challenge 100%
• Reinforcement Learning 100%
• Function Approximation 100%
• Transition Matrix 100%
• on-line algorithm 100%
• State Occupancy 100%