An adaptive deep reinforcement learning framework enables curling robots with human-like performance in real-world conditions

Keyphrases

• Human-like 100%
• Deep Reinforcement Learning (deep RL) 100%
• Reinforcement Learning Framework 100%
• Temporal Features 33%
• Non-stationarity 33%
• Human Performance 33%
• Environmental Characteristics 33%
• Throwing 33%
• Adaptive Framework 33%
• Physics-based 33%
• Wheelchair 33%
• Characteristic Changes 33%
• Artificial Intelligence Systems 33%
• Time-based Rules 33%
• Human Teams 33%
• Rules of the Game 33%
• Official Matches 33%

Computer Science

• Robot 100%
• Learning Framework 100%
• Deep Reinforcement Learning 100%
• Temporal Feature 33%
• Artificial Intelligence 33%

Psychology

• Artificial Intelligence 100%