Theory of linear and integer programming
Theory of linear and integer programming
Merit: A unified framework for routing protocol assessment in mobile AD Hoc networks
Proceedings of the 7th annual international conference on Mobile computing and networking
Ad Hoc Wireless Networks: Protocols and Systems
Ad Hoc Wireless Networks: Protocols and Systems
Operations Research
Inverse optimization in high-speed networks
Discrete Applied Mathematics - Special issue: Algorithmic aspects of communication
MERIT: a scalable approach for protocol assessment
Mobile Networks and Applications
Network topology models for multihop wireless networks
ISRN Communications and Networking
| Hi-index | 0.00 |
In many systems it is an essential part of the operation to find optimal paths in large graphs. In order to understand the system, we need to know the path metric under which the optimal paths are chosen. In many cases, however, no explicite knowledge of the metric is available, we can only observe the path choices that are made. Our goal is to learn the unknown metric, as accurately as possible, from the observed path choices. We present a mathematical model and method to handle this problem. Our main result is that the unknonw path metric can be optimally learned by a polynomial time algorithm, if we assume an additive (but unknown) metric over the graph edges. We also consider the case when the path metric is general, that is, not necessarily additive.