POPCORN: Partially Observed Prediction COnstrained ReiNforcement Learning.
Joseph FutomaMichael C. HughesFinale Doshi-VelezPublished in: CoRR (2020)
Keyphrases
- partially observed
- reinforcement learning
- prediction accuracy
- reinforcement learning algorithms
- state space
- markov decision processes
- optimal policy
- international business
- prediction algorithm
- prediction error
- function approximation
- decision support system
- supervised learning
- prediction model
- recent advances
- dynamic programming
- learning process
- information technology
- multi agent
- artificial intelligence
- learning algorithm