Current position: Senior AI Research Engineer at Huawei. (webpage).
PhD Thesis:The Regression Model of Machine Translation. Koç University, Department of Computer Engineering. August, 2011. (PDF, Presentation).
Publications: bibtex.php, scholar
Abstract:
Machine translation is the task of automatically finding the translation of a source sentence in the target language. Statistical machine translation (SMT) use parallel corpora or bilingual paired corpora that are known to be translations of each other to find a likely translation for a given source sentence based on the observed translations. The task of machine translation can be seen as an instance of estimating the functions that map strings to strings.
Regression based machine translation (RegMT) approach provides a learning framework for machine translation, separating learning models for training, training instance selection, feature representation, and decoding. We use the transductive learning framework for making the RegMT approach computationally more scalable and consider the model building step independently for each test sentence. We develop training instance selection algorithms that not only make RegMT computationally more scalable but also improve the performance of standard SMT systems. We develop better training instance selection techniques than previous work from given parallel training sentences for achieving more accurate RegMT models using less training instances.
We introduce L_1 regularized regression as a better model than L_2 regularized regression for statistical machine translation. Our results demonstrate that sparse regression models are better than L_2 regularized regression for statistical machine translation in predicting target features, estimating word alignments, creating phrase tables, and generating translation outputs. We develop good evaluation techniques for measuring the performance of the RegMT model and the quality of the translations. We use F_1 measure, which performs good when evaluating translations into English according to human judgments. F_1 allows us to evaluate the performance of the RegMT models using the target feature prediction vectors or the coefficients matrices learned or a given SMT model using its phrase table without performing the decoding step, which can be computationally expensive.
Decoding is dependent on the representation of the training set and the features used. We use graph decoding on the prediction vectors represented in n-gram or word sequence counts space found in the training set. We also decode using Moses after transforming the learned weight matrix representing the mappings between the source and target features to a phrase table that can be used by Moses during decoding. We demonstrate that sparse L_1 regularized regression performs better than L_2 regularized regression in the German-English translation task and in the Spanish-English translation task when using small sized training sets. Graph based decoding can provide an alternative to phrase-based decoding in translation domains having low vocabulary.
Full post...
August 16, 2011
July 30, 2011
RegMT System for Machine Translation, System Combination, and Evaluation
Ergun Bicici; Deniz Yuret. Proceedings of the Sixth Workshop on Statistical Machine Translation. pp. 323-329. Edinburgh, Scotland. July, 2011. (PDF, BIB, Proceedings, Poster)
Abstract: We present the results we obtain using our RegMT system, which uses transductive regression techniques to learn mappings between source and target features of given parallel corpora and use these mappings to generate machine translation outputs. Our training instance selection methods perform feature decay for proper selection of training instances, which plays an important role to learn correct feature mappings. RegMT uses L2 regularized regression as well as L1 regularized regression for sparse regression estimation of target features. We present translation results using our training instance selection methods, translation results using graph decoding, system combination results with RegMT, and performance evaluation with the
F1 measure over target features as a metric for evaluating translation quality.
Full post... Related link
Abstract: We present the results we obtain using our RegMT system, which uses transductive regression techniques to learn mappings between source and target features of given parallel corpora and use these mappings to generate machine translation outputs. Our training instance selection methods perform feature decay for proper selection of training instances, which plays an important role to learn correct feature mappings. RegMT uses L2 regularized regression as well as L1 regularized regression for sparse regression estimation of target features. We present translation results using our training instance selection methods, translation results using graph decoding, system combination results with RegMT, and performance evaluation with the
F1 measure over target features as a metric for evaluating translation quality.
Full post... Related link
Instance Selection for Machine Translation using Feature Decay Algorithms
Ergun Bicici; Deniz Yuret. Proceedings of the Sixth Workshop on Statistical Machine Translation. pp. 272-283. Edinburgh, Scotland. July, 2011. (PDF, BIB, Proceedings, Presentation)
Abstract: We present an empirical study of instance selection techniques for machine translation. In an active learning setting, instance selection minimizes the human effort by identifying the most informative sentences for translation. In a transductive learning setting, selection of training instances relevant to the test set improves the final translation quality. After reviewing the state of the art in the field, we generalize the main ideas in a class of instance selection algorithms that use feature decay. Feature decay algorithms increase diversity of the training set by devaluing features that are already included. We show that the feature decay rate has a very strong effect on the final translation quality whereas the initial feature values, inclusion of higher order features, or sentence length normalizations do not. We evaluate the best instance selection methods using a standard Moses baseline using the whole 1.6 million sentence English-German section of the Europarl corpus. We show that selecting the best 3000 training sentences for a specific test sentence is sufficient to obtain a score within 1 BLEU of the baseline, using 5% of the training data is sufficient to exceed the baseline, and a ~ 2 BLEU improvement over the baseline is possible by optimally selected subset of the training data. In out-of-domain translation, we are able to reduce the training set size to about 7% and achieve a similar performance with the baseline.
Full post... Related link
Abstract: We present an empirical study of instance selection techniques for machine translation. In an active learning setting, instance selection minimizes the human effort by identifying the most informative sentences for translation. In a transductive learning setting, selection of training instances relevant to the test set improves the final translation quality. After reviewing the state of the art in the field, we generalize the main ideas in a class of instance selection algorithms that use feature decay. Feature decay algorithms increase diversity of the training set by devaluing features that are already included. We show that the feature decay rate has a very strong effect on the final translation quality whereas the initial feature values, inclusion of higher order features, or sentence length normalizations do not. We evaluate the best instance selection methods using a standard Moses baseline using the whole 1.6 million sentence English-German section of the Europarl corpus. We show that selecting the best 3000 training sentences for a specific test sentence is sufficient to obtain a score within 1 BLEU of the baseline, using 5% of the training data is sufficient to exceed the baseline, and a ~ 2 BLEU improvement over the baseline is possible by optimally selected subset of the training data. In out-of-domain translation, we are able to reduce the training set size to about 7% and achieve a similar performance with the baseline.
Full post... Related link
June 21, 2011
ACL 2011 Tutorials
Here are some notes from the ACL tutorials:
Jeffrey Heinz, Colin de la Higuera and Menno van Zaanen
Jeffrey and Colin motivated and presented the main results for formal grammatical inference. Even though many theoretical results are negative, learning is usually possible by restricting the model class (to a well defined subset used by natural languages) or assuming a non-distribution-free setting. Colin's book was recommended as a good introduction to the theory. It would be interesting to see if Turkish morphotactics or morphophonemics fall into one of the easy-to-learn model subclasses. You can download the slides.
Rich Prior Knowledge in Learning for Natural Language Processing
Gregory Druck, Kuzman Ganchev, Joao Graca
Gregory Druck, Kuzman Ganchev, Joao Graca
There is a recent trend for encoding prior knowledge in learning problems not in the prior distributions but later in the learning process. The prior knowledge usually comes in the form of feature-class expectations and guiding the model toward the correct expectations is only possible after considering the input. Posterior regularization, constraint driven learning, and generalized expectation criteria seem to be related implementations of this idea. You can download the slides.
Dual Decomposition for Natural Language Processing
Michael Collins and Alexander M Rush
Michael Collins and Alexander M Rush
I did not attend this one but here are the slides.
Full post...
May 23, 2011
Semantic Structures by Ray Jackendoff
Jackendoff divides the study of the language faculty into three components: phonological, syntactic, and conceptual. In speech recognition a model that tells us what word sequences are likely is essential to recognition accuracy, because the acoustic information is often ambiguous (e.g. "She kissed this guy." vs. "She kissed the sky."). Stated in Jackendoff's terms, a probabilistic model of the syntactic component is helpful in disambiguating what is going on in the phonological component. Similarly I think a model of what is likely in the conceptual component is essential to resolving ambiguities in the syntactic component. If "what is likely to be said" is essential in interpreting "what is heard", then "what is likely to be meant" is similarly essential in interpreting "what is said".
Unfortunately we do not have good conceptual models yet, so computational linguists still try to make do with error prone hand tagging and shallow machine learning to disambiguate senses, references, and relations.
On a side note, each component in Jackendoff's work is modeled after the generative paradigm which, for the syntactic component, is described as follows:
Am I the only one befuddled by this argument? Primitives plus means of combination is certainly one way to create infinity using finite means, but why assume it is the only way? Dynamic systems, random processes, who knows what else can lead to infinite possible outcomes from a finite initial endowment. Why just present two strawmen, finite and infinite lists, as the only alternatives to discrete primitives and combination? Why after a couple of paragraphs further narrow the description to "the argument from creativity to the NECESSITY for principles or rules in syntactic knowledge"? Discrete primitives with finite and definite constraints and rules of combination is one way to build a representational system, unlikely to be the correct way for all three components of language, and certainly not the only way.
See also: Plausibility vs. Inference.
Full post... Related link
Unfortunately we do not have good conceptual models yet, so computational linguists still try to make do with error prone hand tagging and shallow machine learning to disambiguate senses, references, and relations.
On a side note, each component in Jackendoff's work is modeled after the generative paradigm which, for the syntactic component, is described as follows:
- Speakers can understand and create an indefinite number of sentences they have never heard before.
- Therefore the repertoire of syntactic structures cannot be characterized as a finite list of sentences.
- Nor can it be characterized as an infinite list of possible sentences because we have finite brains.
- Thus it MUST be mentally encoded in terms of a finite set of primitives and a finite set of principles of combination that collectively generate the class of possible sentences.
Am I the only one befuddled by this argument? Primitives plus means of combination is certainly one way to create infinity using finite means, but why assume it is the only way? Dynamic systems, random processes, who knows what else can lead to infinite possible outcomes from a finite initial endowment. Why just present two strawmen, finite and infinite lists, as the only alternatives to discrete primitives and combination? Why after a couple of paragraphs further narrow the description to "the argument from creativity to the NECESSITY for principles or rules in syntactic knowledge"? Discrete primitives with finite and definite constraints and rules of combination is one way to build a representational system, unlikely to be the correct way for all three components of language, and certainly not the only way.
See also: Plausibility vs. Inference.
Full post... Related link
Subscribe to:
Comments (Atom)
