Difference between revisions of "Margin Infused Relaxed Algorithm"
From Cohen Courses
Jump to navigationJump to search| Line 2: | Line 2: | ||
== Summary == | == Summary == | ||
| − | MIRA is an online large-margin training algorithm which updates the weight vector according to certain margin constraints and loss function. It is used to learn the weights of features after processing each training instance. | + | MIRA is an online large-margin training algorithm which updates the weight vector according to certain margin constraints and loss function. It is used to learn the weights of features after processing each training instance similar to [[UsesMethod:structured perceptron algorithm with an additional loss function in its update rule. |
== Definition == | == Definition == | ||
Revision as of 18:28, 29 September 2011
This method is used by Watanabe et al., EMNLP 2007 to train an MT system a with a very large number of features of the order of millions. The training step was performed using a specific algorithm called the Margin Infused Relaxed Algorithm (MIRA) proposed by
Summary
MIRA is an online large-margin training algorithm which updates the weight vector according to certain margin constraints and loss function. It is used to learn the weights of features after processing each training instance similar to [[UsesMethod:structured perceptron algorithm with an additional loss function in its update rule.
