Rotting bandits are no harder than stochastic ones.
Julien SeznecAndrea LocatelliAlexandra CarpentierAlessandro LazaricMichal ValkoPublished in: CoRR (2018)
Keyphrases
- stochastic systems
- stochastic models
- regret bounds
- database
- monte carlo
- np complete
- multi armed bandit
- multiscale
- stochastic model
- stochastic processes
- learning algorithm
- stochastic nature
- objective function
- expert systems
- np hard
- decision trees
- artificial intelligence
- state transition
- learning automata
- stochastic approximation
- machine learning
- neural network
- databases