Thompson Sampling For Stochastic Bandits with Graph Feedback.
Aristide C. Y. TossouChristos DimitrakakisDevdatt P. DubhashiPublished in: CoRR (2017)
Keyphrases
- pairwise
- graph matching
- stochastic systems
- monte carlo
- multi armed bandit
- graph structure
- weighted graph
- graph theory
- random sampling
- random walk
- directed graph
- sampling strategy
- learning automata
- graph representation
- bipartite graph
- directed acyclic graph
- sampling methods
- graph construction
- sample size
- relevance feedback
- neural network
- confidence intervals
- graph theoretic
- structured data
- regret bounds
- graphical models
- metropolis hastings