I'm a doctoral student interested in machine translation, unreasonably effective neural networks, phonetics and acoustics, markup technologies. Also, I find teaching super exciting!
Planning a programming project or thesis related to the topics mentioned above? Come talk to me, I'd love to hear about it.
What I am currently doing
I am working with Annette Rios and Rico Sennrich on integrating linguistic information into current neural machine translation systems. I am part of an SNF project, CONTRA, which is led by Martin Volk.
In the particular area of neural machine translation I am most interested in the following:
- Augmenting standard bidirectional RNN encoder-decoder-attention architectures with additional context.
- Non-recurrent models for neural machine translation, such as fully convolutional or self-attentional (transformer) models.
- Contrastive evaluation of NMT systems.
- Dual learning of two translation directions and reinforcement learning.
- Pre-training of NMT models with language models, without recurrent structures.
- Deep contextualized embeddings
- Defensive distillation to guard against adversarial attacks
- Domain robustness of neural MT systems
(Also, I very frequently change my mind about "most interested in" - so this list is growing uncontrollably, sorry!)
Things I used to do
- Making people shout with controlled loudness, with Lei He and Kostis Dimos.
- Investigating rhythmic variability in speakers of many languages, anything from Zurich Swiss German to Burmese, with Volker Dellwo.
- Developing the statistical machine translation tool mtrain, with Samuel Läubli.
- Being very active in all things XML on Stackoverflow, with Kenneth Hughes and learning the ropes from Michael Kay.
|2018||Quantitative Methods and Machine Learning, with Manfred Klenner|
|2018||XML, Informationsmanagement, with Tobias Hodel and Jeannette Isele|
|2018||Machine Translation, with Samuel Läubli|
|2017||Quantitative Methods and Machine Learning, with Manfred Klenner|
|2017||Teaching Assistant for Machine Translation and Parallel Corpora, with Martin Volk|
|2016||Teaching Assistant for XML Technologies and Semantic Web, with Fabio Rinaldi|
|2015||Teaching Assistant for Machine Translation and Parallel Corpora, with Martin Volk|
|2014||Teaching Assistant for XML Technologies and Semantic Web, with Fabio Rinaldi|
|2012||Teaching Assistant for Sprachtechnologie für Barrierefreiheit, with Sarah Ebling|
Tang, Gongbo; Müller, Mathias; Rios, Annette; Sennrich, Rico (2018). Why Self-Attention? A Targeted Evaluation of Neural Machine Translation Architectures. In: Conference on Empirical Methods in Natural Language Processing (EMNLP), Brussels/Belgium, 31 October 2018 - 5 November 2018.
Müller, Mathias; Rios, Annette; Voita, Elena; Sennrich, Rico (2018): A Large-Scale Test Set for the Evaluation of Context-Aware Pronoun Translation. In: Third Conference on Machine Translation (WMT), Brussels/Belgium, 31 October 2018 - 1 November 2018.
Rios, Annette; Müller, Mathias; Sennrich, Rico (2018): The Word Sense Disambiguation Test Suite at WMT18. In: Third Conference on Machine Translation (WMT), Brussels/Belgium, 31 October 2018 - 1 November 2018.
Läubli, Samuel; Müller, Mathias; Horat, Beat; Volk, Martin (2018). mtrain: A Convenience Tool for Machine Translation. In: Annual Conference of the European Association for Machine Translation (EAMT), Alicante/Spain, 28 May 2018 - 30 May 2018.
Müller, Mathias (2017). Treatment of Markup in Statistical Machine Translation. In: Third Workshop on Discourse in Machine Translation, EMNLP, Copenhagen/Denmark, 8 September 2017.
Volk, Martin; Amrhein, Chantal; Aepli, Noëmi; Müller, Mathias; Ströbel, Phillip (2016). Building a Parallel Corpus on the World's Oldest Banking Magazine. In: KONVENS, Bochum/Germany, 19 September 2016 - 21 September 2016.
Dimos, Kostis; He, Lei; Dellwo, Volker; Müller, Mathias (2015). A method for the elicitation of shouted speech with controlled loudness. In: IAFPA Annual Conference, Leiden/Netherlands, 8 July 2015 - 10 July 2015.
Müller, Mathias; Volk, Martin (2013). Statistical machine translation of subtitles: From OpenSubtitles to TED. In: Gurevych, Iryna; Biemann, Chris; Zesch, Torsten. Language Processing and Knowledge in the Web. Berlin Heidelberg: Springer, 132-138.
Dellwo, Volker; Schmid, Stephan; Leemann, Adrian; Kolly, Marie-José; Müller, Mathias (2012). Speaker identification based on speech rhythm: the case of bilinguals. In: Perspectives on Rhythm and Timing (PoRT), Glasgow/Scotland, 19 July 2012 - 21 July 2012.
Talks / Posters
Müller, Mathias (2018). Neural Machine Translation. Presented at: BrainHack Zurich, Zurich/Switzerland, 3 May 2018.
Müller, Mathias (2017). Treatment of Markup in Statistical Machine Translation. Presented at: GSCL, Berlin/Germany, 13 September 2017.
Müller, Mathias (2016). Schreien im Labor - Sprecheridentifizierung und forensische Phonetik. Presented at: Weiterbildungsseminar Akustika, Olten/Switzerland, 8 November 2016.
Nespeca, Roberto; Müller, Mathias (2016). Integrating a customized SMT system into traditional computed-aided translation workflows. Presented at: Swisstext, Winterthur/Switzerland, 8 June 2016.
Schmid, Stephan; Dellwo, Volker; Tramutoli, Laura; Müller, Mathias (2011). Is the speech rhythm of bilinguals native or non-native? Presented at: Prosodic and rhythmic aspects of L2 Italian, Naples/Italy, 5 May 2011 - 6 May 2011.