LEARNING LONG-TERM REWARD REDISTRIBUTION VIA RANDOMIZED RETURN DECOMPOSITION

Keyphrases

• Reward Redistribution 100%
• Return Decomposition 100%
• Long-term Reward 100%
• Reward Signal 66%
• Episodic Reinforcement Learning 66%
• Popular 33%
• Problem Formulation 33%
• Least Squares 33%
• Environmental Signals 33%
• Reward Function 33%
• Future Outcomes 33%
• Reward Delay 33%
• Monte Carlo Sampling 33%
• Baseline Algorithm 33%
• Redistribution Algorithm 33%
• Algorithmic Properties 33%
• Surrogate Loss Function 33%
• Sparse Reward 33%
• Episodic Rewards 33%
• Benchmark Tasks 33%
• Delayed Reward 33%
• Reinforcement Learning Applications 33%
• Dense Reward 33%

Computer Science

• Reinforcement Learning 100%
• Least Squares Method 33%
• Problem Formulation 33%
• Randomized Algorithm 33%
• Problem Setting 33%
• Benchmark Task 33%