Cornering Stationary and Restless Mixing Bandits with Remix-UCB.
Julien AudiffrenLiva RalaivolaPublished in: NIPS (2015)
Keyphrases
- multi armed bandit
- multi armed bandits
- non stationary
- stochastic systems
- semi markov
- conservation laws
- reinforcement learning
- optimal control
- bandit problems
- blind source separation
- decision problems
- information retrieval
- data sets
- real time
- expert systems
- video sequences
- regret bounds
- similarity measure
- machine learning
- neural network