Jacobs University Bremen

About: Jacobs University Bremen is a education organization based out in Bremen, Germany. It is known for research contribution in the topics: Population & Galaxy cluster. The organization has 2184 authors who have published 6435 publications receiving 245199 citations. The organization is also known as: International University Bremen & IUB.

Neural Machine Translation by Jointly Learning to Align and Translate

Abstract: Neural machine translation is a recently proposed approach to machine translation. Unlike the traditional statistical machine translation, the neural machine translation aims at building a single neural network that can be jointly tuned to maximize the translation performance. The models proposed recently for neural machine translation often belong to a family of encoder-decoders and consists of an encoder that encodes a source sentence into a fixed-length vector from which a decoder generates a translation. In this paper, we conjecture that the use of a fixed-length vector is a bottleneck in improving the performance of this basic encoder-decoder architecture, and propose to extend this by allowing a model to automatically (soft-)search for parts of a source sentence that are relevant to predicting a target word, without having to form these parts as a hard segment explicitly. With this new approach, we achieve a translation performance comparable to the existing state-of-the-art phrase-based system on the task of English-to-French translation. Furthermore, qualitative analysis reveals that the (soft-)alignments found by the model agree well with our intuition.

The SILVA ribosomal RNA gene database project: improved data processing and web-based tools

Abstract: SILVA (from Latin silva, forest, http://www.arb-silva.de) is a comprehensive web resource for up to date, quality-controlled databases of aligned ribosomal RNA (rRNA) gene sequences from the Bacteria, Archaea and Eukaryota domains and supplementary online services. The referred database release 111 (July 2012) contains 3 194 778 small subunit and 288 717 large subunit rRNA gene sequences. Since the initial description of the project, substantial new features have been introduced, including advanced quality control procedures, an improved rRNA gene aligner, online tools for probe and primer evaluation and optimized browsing, searching and downloading on the website. Furthermore, the extensively curated SILVA taxonomy and the new non-redundant SILVA datasets provide an ideal reference for high-throughput classification of data from next-generation sequencing approaches.

Neural Machine Translation by Jointly Learning to Align and Translate

Abstract: Neural machine translation is a recently proposed approach to machine translation. Unlike the traditional statistical machine translation, the neural machine translation aims at building a single neural network that can be jointly tuned to maximize the translation performance. The models proposed recently for neural machine translation often belong to a family of encoder-decoders and consists of an encoder that encodes a source sentence into a fixed-length vector from which a decoder generates a translation. In this paper, we conjecture that the use of a fixed-length vector is a bottleneck in improving the performance of this basic encoder-decoder architecture, and propose to extend this by allowing a model to automatically (soft-)search for parts of a source sentence that are relevant to predicting a target word, without having to form these parts as a hard segment explicitly. With this new approach, we achieve a translation performance comparable to the existing state-of-the-art phrase-based system on the task of English-to-French translation. Furthermore, qualitative analysis reveals that the (soft-)alignments found by the model agree well with our intuition.

Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies

Abstract: 16S ribosomal RNA gene (rDNA) amplicon analysis remains the standard approach for the cultivation-independent investigation of microbial diversity. The accuracy of these analyses depends strongly on the choice of primers. The overall coverage and phylum spectrum of 175 primers and 512 primer pairs were evaluated in silico with respect to the SILVA 16S/18S rDNA non-redundant reference dataset (SSURef 108 NR). Based on this evaluation a selection of 'best available' primer pairs for Bacteria and Archaea for three amplicon size classes (100-400, 400-1000, ≥ 1000 bp) is provided. The most promising bacterial primer pair (S-D-Bact-0341-b-S-17/S-D-Bact-0785-a-A-21), with an amplicon size of 464 bp, was experimentally evaluated by comparing the taxonomic distribution of the 16S rDNA amplicons with 16S rDNA fragments from directly sequenced metagenomes. The results of this study may be used as a guideline for selecting primer pairs with the best overall coverage and phylum spectrum for specific applications, therefore reducing the bias in PCR-based microbial diversity studies.

On the Properties of Neural Machine Translation: Encoder--Decoder Approaches

Abstract: Neural machine translation is a relatively new approach to statistical machine translation based purely on neural networks. The neural machine translation models often consist of an encoder and a decoder. The encoder extracts a fixed-length representation from a variable-length input sentence, and the decoder generates a correct translation from this representation. In this paper, we focus on analyzing the properties of the neural machine translation using two models; RNN Encoder‐Decoder and a newly proposed gated recursive convolutional neural network. We show that the neural machine translation performs relatively well on short sentences without unknown words, but its performance degrades rapidly as the length of the sentence and the number of unknown words increase. Furthermore, we find that the proposed gated recursive convolutional network learns a grammatical structure of a sentence automatically.