Most statistical machine translation (SMT) systems are modeled using a log-linear framework.

Although the log-linear model achieves success in SMT, it still suffers from some limitations: (1) the features are required to be linear with respect to the model itself; (2) features cannot be further interpreted to reach their potential.

additive neural networks, for SMT to go beyond the log-linear translation model.

Recently, great progress has been achieved in SMT, especially since Och and Ney (2002) proposed the log-linear model: almost all the state-of-the-art SMT systems are based on the log-linear model.

Regardless of how successful the log-linear model is in SMT, it still has some shortcomings.

Compared with the log-linear model, it has more powerful expressive abilities and can deeply interpret and represent features with hidden units in neural networks.

As is standard in such settings, the distribution will be log-linear in a set of features of these parses.

Instead, we assume that the distribution over yCFG is a log-linear model with parameters 601:0 (i.e., a sub-vector of 6) , namely:

In this setup, the model reduces to a normal log-linear model for the target formalism.

It’s not surprising that Cahill’s model outperforms our log-linear model because it relies heavily on handcrafted rules optimized for the dataset.

Feature functions in log-linear models are designed to capture the characteristics of each derivation in the tree.

log-linear models with parameters, AiméRM , is

trained to predict Blgikfik, for each w, of a tag 81.32 BEE) = exp(f(wl; A?» (2) where the log-linear function f is: n23} = m; A295 = 9531905771 (3)

labeled data, 712?, based on the log-linear model in Eq.

The a: is used as the input matrix of the kth log-linear model (corresponding to kth semantic tag (topic)) to infer the [3 hyper-parameter of MTR in Eq.

We then report the corresponding chains 0(a) as the system output.3 For learning, the gradient takes the standard form of the gradient of a log-linear model, a difference of expected feature counts under the gold annotation and under no annotation.

We use a log-linear model that can be expressed as a factor graph.

Each unary factor A,- has a log-linear form with features examining mention 2', its selected antecedent ai, and the document context at.

The final log-linear model is given by the following formula:

(4) The orientation probability is modeled in a log-linear framework using a set of N feature func-tiOIlS €{,i,j, de-IJH), n = 1, .

Finally, in the log-linear framework (Equation 2) a new jump model is added which uses the reordered source sentence to calculate the cost.

The number of source words that have inconsistent labels is the penalty and is then added into the log-linear framework as a new feature.

We model Pr(e{|fi]) directly using a log-linear combination of several models (Och and Ney,

Our translation model is embedded in a log-linear model as is common for SMT, and treated as a single translation model in this log-linear combination.

Log-linear weights are optimized using MERT (Och and Ney, 2003).

Future work could involve merging our translation model framework with the online adaptation of other models, or the log-linear weights.

The noise model that gbc parameterizes is a local log-linear model, so we follow the approach of Berg-Kirkpatrick et al.

logistic( Z [M917 kw - Wedgie/D k’ :1 The fact that the parameterization is log-linear will ensure that, during the unsupervised learning process, updating the shape parameters gbc is simple and feasible.

(2010), we use a regularization term in the optimization of the log-linear model parameters (15¢ during the M-step.

One is the log-linear combination of TMs trained on each subcorpus (Koehn and Schroeder, 2007), with weights of each model tuned under minimal error rate training using MIRA.

Research on mixture models has considered both linear and log-linear mixtures.

(Koehn and Schroeder, 2007), instead, opted for combining the sub-models directly in the SMT log-linear framework.

Our phrase-based decoder is similar to Moses (Koehn et al., 2007) and uses the phrase pairs and target LM to perform beam search stack decoding based on a standard log-linear model, the parameters of which were tuned with MERT (Och, 2003) on a held-out development set (3,534 sentence pairs, 45K words) using BLEU as the tuning metric.

We add this feature to the log-linear translation model with its own weight, which is tuned with MERT.

Translation phrase pairs that originate in training conversations whose topic distribution is similar to that of the current conversation are given preference through a single similarity feature, which augments the standard phrase-based SMT log-linear model.

Given a single instruction 2' with category au, we use a log-linear model to represent the distri-

We employ a log-linear model and try to maximize initial dialog state distribution over the space of all nodes in a dialog network:

Dialog State Update Model We use a log-linear model to maximize a dialog state distribution over the space of all nodes in a dialog network:

But instead of using just the PMI scores of bilingual NE pairs, as in our work, they employed a feature-rich log-linear model to capture bilingual correlations.

Parameters in their log-linear model require training with bilingually annotated data, which is not readily available.

(2010a) presented a supervised learning method for performing joint parsing and word alignment using log-linear models over parse trees and an ITG model over alignment.