Expansive Latent Planning for Sparse Reward Offline Reinforcement Learning.
Robert GieselmannFlorian T. PokornyPublished in: CoRL (2023)
Keyphrases
- reinforcement learning
- action selection
- partially observable
- state space
- macro actions
- function approximation
- deterministic domains
- complex domains
- partially observable markov decision processes
- heuristic search
- reinforcement learning algorithms
- partially observable environments
- eligibility traces
- reward shaping
- reward function
- markov decision problems
- learning algorithm
- stochastic domains
- sparse data
- model free
- optimal policy
- planning problems
- temporal difference
- reinforcement learning methods
- learning agent
- markov decision processes
- latent variables
- motion planning
- learning problems
- compressive sensing
- average reward
- ai planning
- transfer learning
- machine learning
- domain independent
- planning systems
- control policy
- sparse representation
- dynamic programming
- learning process
- multi agent systems