A Structured Multiarmed Bandit Problem and the Greedy Policy.
Adam J. MersereauPaat RusmevichientongJohn N. TsitsiklisPublished in: IEEE Trans. Autom. Control. (2009)
Keyphrases
- multiarmed bandit
- greedy algorithm
- structured data
- search algorithm
- optimal policy
- locally optimal
- machine learning
- decision process
- database
- information extraction
- dynamic programming
- real world
- evolutionary algorithm
- website
- feature selection
- long run
- expected cost
- asymptotically optimal
- markov decision process
- greedy algorithms
- policy making
- management policies
- forward selection
- neural network