One Step at a Time: Pros and Cons of Multi-Step Meta-Gradient Reinforcement Learning.
Clément BonnetPaul CaronThomas BarrettIan DaviesAlexandre LaterrePublished in: CoRR (2021)
Keyphrases
- multi step
- pros and cons
- reinforcement learning
- single step
- lower bounding
- knn
- k nearest neighbor
- temporal difference
- tumor classification
- distance computation
- function approximation
- td learning
- reinforcement learning algorithms
- markov decision processes
- semi supervised
- state space
- policy gradient
- data mining
- nearest neighbor
- feature vectors
- learning algorithm