Hamidreza Ghader

PhD student of Computer Science.
Information and Language Processing Systems (ILPS)
Informatics Institute (IvI)
University of Amsterdam (UvA)

Research Interests
Neural Machine Translation
Statistical Machine Translation
Natural Language Processing
Statistical Machine Learning

Detailed CV


Address
Office C3.234
Science Park 904
1098 XH Amsterdam
The Netherlands

Email
h dot surname at uva dot nl

I am a PhD candidate at University of Amsterdam in the Information and Language Processing Systems (ILPS) group. I am working on Statistical Machine Translation with Dr. Christof Monz. Before starting my PhD, I was working as a research assistant in Natural Language and Text Processing Laboratory at University of Tehran. There, we were developing Faraazin machine translation system.

Short Bio

BSc. in Computer Engineering at ECE Department, University of Tehran.

MSc. in Artificial Intelligence at Computer Engineering Department, Iran University of Science and Technology

Selected Works


What does Attention in neural machine translation pay attention to?

Attention in neural machine translation provides the possibility to encode relevant parts of the source sentence at each translation step. As a result, attention is considered to be an alignment model as well. However, there is no work that specifically studies attention and provides analysis of what is being learned by attention models. Thus, the question still remains that how attention is similar or different from the traditional alignment. In this paper, we provide detailed analysis of attention and compare it to traditional alignment. We answer the question of whether attention is only capable of modelling translational equivalent or it captures more information. We show that attention is different from alignment in some cases and is capturing useful information other than alignments.

More Info

Which Words Matter in Defining Phrase Reorderings in Statistical Machine Translation?

We propose two models to use shorter sub-phrase pairs of an original phrase pair to smooth the phrase reordering distributions. In the first model we follow the classic idea of backing off to shorter histories commonly used in language model smoothing. In the second model, we use syntactic dependencies to identify the most relevant words in a phrase to back off to. We show how these models can be easily applied to existing lexicalized and hierarchical reordering models. The results show that not all the words inside a phrase pair are equally important in defining phrase reordering behavior and shortening towards important words will decrease the sparsity problem for long phrase pairs.

More Info

Automatic WordNet Construction Using Markov chain Monte Carlo

In this work we proposed a fully-automated approach for constructing a Persian WordNet. Our acquired WordNet has a precision of 90.46% which is a considerable improvement in comparison with automatically-built WordNets in Persian. Just send me an email if you want the WordNet.

More Info

Faraazin Machine Translation System

Formerly, I was a member of machine translation development team in Natural Language and Text Processing Laboratory at University of Tehran. There we developed a hybrid machine translation system which combines transfer-based models with statistical approaches.

More Info