Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
| Publication date | 2012 |
|---|---|
| Host publication | Network Games, Control and Optimization (NetGCooP), 2012 6th International Conference on |
| Place of Publication | New York |
| Publisher | IEEE |
| Pages | 38-43 |
| Number of pages | 6 |
| ISBN (print) | 9781467360265 |
| <mark>Original language</mark> | English |
| Event | 6th International Conference on Network Games, Control and OPtimization (NetGCooP) - Avignon, France Duration: 28/11/2012 → 30/11/2012 |
| Conference | 6th International Conference on Network Games, Control and OPtimization (NetGCooP) |
|---|---|
| Country/Territory | France |
| Period | 28/11/12 → 30/11/12 |
| Conference | 6th International Conference on Network Games, Control and OPtimization (NetGCooP) |
|---|---|
| Country/Territory | France |
| Period | 28/11/12 → 30/11/12 |
We consider Content Centric Network (CCN) interest forwarding problem as a Multi-Armed Bandit (MAB) problem with delays. We investigate the transient behaviour of the epsilon-greedy, tuned epsilon-greedy and Upper Confidence Bound (UCB) interest forwarding policies. Surprisingly, for all the three policies very short initial exploratory phase is needed. We demonstrate that the tuned epsilon-greedy algorithm is nearly as good as the UCB algorithm, commonly reported as the best currently available algorithm. We prove the uniform logarithmic bound for the tuned epsilon-greedy algorithm in the presence of delays. In addition to its immediate application to CCN interest forwarding, the new theoretical results for MAB problem with delays represent significant theoretical advances in machine learning discipline.