BOOK: Storing Algorithm-Invariant Episodes for Deep Reinforcement Learning.
Simyung ChangYoungjoon YooJaeseok ChoiNojun KwakPublished in: ICPRAM (2019)
Keyphrases
- dynamic programming
- reinforcement learning
- learning algorithm
- improved algorithm
- linear programming
- experimental evaluation
- detection algorithm
- computational complexity
- significant improvement
- preprocessing
- probabilistic model
- computational cost
- machine learning
- times faster
- simulated annealing
- worst case
- optimal solution
- objective function
- book presents
- cost function
- neural network
- researchers and practitioners
- recognition algorithm
- convergence rate
- image segmentation
- tree structure
- clustering method
- segmentation algorithm
- theoretical analysis
- expectation maximization
- multi agent
- high accuracy
- k means