Bridging the gap between Markowitz planning and deep reinforcement learning.
Eric BenhamouDavid SaltielSandrine UngariAbhishek MukhopadhyayPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- action selection
- state space
- macro actions
- learning algorithm
- partially observable
- ai planning
- model free
- deterministic domains
- domain independent
- planning process
- reinforcement learning algorithms
- heuristic search
- machine learning
- stochastic domains
- partial observability
- function approximation
- planning problems
- markov decision processes
- temporal difference
- decision support
- complex domains
- motion planning
- policy search
- optimal policy
- evolutionary algorithm
- search space
- multi agent
- deep learning
- portfolio selection
- multi agent reinforcement learning
- reinforcement learning problems