J12-1003 improves performance compared with a greedy optimization procedure . Table 4 demonstrates
P03-1015 for some combinations with the greedy optimization strategy on ideal tags . All
D12-1006 . Alternatively , we can use a greedy optimization algorithm to find partitions
P10-1084 kernelized input vectors . We use greedy optimization during training based on ROUGE
P03-1015 dependent on the POS tag as well . For greedy optimization , the predictions of the finite-state
D12-1006 that : C * = arg min P ( C ) . A greedy optimization framework is used to minimize
E14-1075 how summaries are generated by a greedy optimization system which selects the sentence
N12-3009 - rithms . The first one is a greedy optimization algorithm that is based on the
P09-1031 optimization can be solved by a greedy optimization algorithm . At each term insertion
P03-1015 results of the base classifiers . Greedy optimization outperforms all other strategies
P11-1052 means that an efficient scalable greedy optimization scheme has a constant factor
P03-1015 were assigned multiple values . Greedy Optimization of F-value . Another method uses
P03-1015 classifier . The list is obtained by greedy optimization : In each step , the prediction
P10-1041 a local optimum because of its greedy optimization flavor . Kamps et al. ( 2004
J14-3003 a local optimum because of its greedy optimization flavor . Kamps et al. ( 2004
P03-1015 learning , Maximum Entropy , and greedy optimization of F-value . Simple Voting .
D12-1120 and we develop effi - cient , greedy optimization techniques for learning effective
J12-1003 ILP-Global . Thus , we see that greedy optimization might get stuck in local maxima
E14-1037 optimization problem . We use a greedy optimization procedure that iteratively updates
N04-1004 multimodal communication . Using greedy optimization and only a minimum of linguistic
hide detail