Network flows: theory, algorithms, and applications
Network flows: theory, algorithms, and applications
A simple rule-based part of speech tagger
ANLC '92 Proceedings of the third conference on Applied natural language processing
Sequential and parallel algorithms for minimum flows
The Korean Journal of Computational & Applied Mathematics
Neuro-evolution approaches to collective behavior
CEC'09 Proceedings of the Eleventh conference on Congress on Evolutionary Computation
Collective Behavior of a Small-World Recurrent Neural System With Scale-Free Distribution
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
The article presents a preflow algorithm for the parametric minimum flow problem working in a parametric residual network with linear lower bound functions of a single parameter. On each of the iterations, the highest-label partitioning preflow-pull algorithm (H-L P P-P) pulls flow from an active node with the highest distance label over a conditionally admissible arc. After each pull of flow, either the parametric residual capacity of the arc or the parametric deficit of the node becomes zero for at least a subinterval of the parameter values. If the two situations take place in different subintervals, the algorithm is continued in two different parametric residual networks generated by this partitioning pull. The template-like structure of a dialogue act reveals a design where information about the items (part-of-speech) is a multiple section vector with one segment for each of the used part of speech categories. These categories are divided into groups, according to their importance regarding the task, enabling each segment to use its own representations for the words within it.