Natalie Schluter

About

My main topics of interest are in Theoretical Computer Science (Graph Algorithms, Automata and Machine Learning) and applications for these theoretical findings in Natural Language Processing and other Data Science.

Current Position:
I am currently Machine Learning Researcher at Apple in Samy Bengio's super exciting new Machine Learning Research Organisation.

I am also Associate Professor at the Department of Applied Mathematics and Computer Science, Technical University of Denmark.

In March 2020, I stepped down as Head of Programme for the new Data Science BSc at the IT University of Copenhagen.  Together with some brilliant colleagues I started this programme up. The first students arrived in August 2017. And it was a blast being a part of that craziness.

Former Positions:

• Senior Research Scientist Google Brain (now Google Deep Mind).
• Chief Analyst and Lead Data Scientist at MobilePay by Danske Bank, Copenhagen, Denmark
• Postdoctoral researcher in Natural Language Processing, the University of Copenhagen, Denmark
• Postdoctoral researcher in Data Science and Natural Language Processing at Malmö University, Sweden

Previous Education:

• PhD in Natural Language Processing (Computer Science) from Dublin City University, Ireland
  • (also with PhD Studies in Graph Algorithms at IT University of Copenhagen)
• MSc in Mathematics from Trinity College, University of Dublin, Ireland
• MA in Linguistics (Semantics) from Université de Montréal, Canada
• Honours BSc in Mathematics from Université de Montréal, Canada
• BA in French and Spanish from the University of British Columbia, Vancouver, Canada

Other stuff:

• I am originally from Canada, but now both Danish and Canadian.

Current Students and Research Assistants

• David Sasu, PhD student
• Tsedeniya Kinfe Temesgen, Research Assistant

I no longer supervise new students. Please do check out my colleagues at DTU Compute for supervision possibilities.

Publications

2021

MassiveSumm: a very large-scale, very multilingual, newswire summarisation dataset
Daniel Varab and Natalie Schluter
In Proc of EMNLP, Punta Cana, Dominican Republic.

2020

DaNewsroom: A Large-scale Danish Summarisation Dataset
Daniel Varab and Natalie Schluter
In Proc of LREC, Marseille, France.

2019

On approximating dropout noise injection (arxiv pre-print)
Natalie Schluter

UniParse: A universal graph-based parsing toolkit
Daniel Varab and Natalie Schluter
In Proc. of NoDaLiDa, Turku, Finland.pdf.

The Lacunae of Danish Natural Language Processing
Andreas Kirkedal, Barbara Plank, Leon Derczynski and Natalie Schluter
In Proc of NoDaLiDa, Turku, Finland. pdf.

Recurrent models and lower bounds for projective syntactic decoding
Natalie Schluter
In Proc of NAACL, Minneapolis, Minnesota, USA. pdf

2018

The Glass Ceiling in NLP
Natalie Schluter
In Proc of EMNLP, Brussels, Belgium. pdf

When data permutations are pathological: the case of neural natural language inference
Natalie Schluter and Daniel Varab
In Proc of EMNLP, Brussels, Belgium. pdf

The Word Analogy Testing Caveat
Natalie Schluter
In Proc of NAACL, New Orleans, Louisiana, USA. pdf

Baselines and test data for cross-lingual inference
Željko Agić and Natalie Schluter
In Proc of LREC, Miyazaki, Japan. pdf

2017

The limits of automatic summarisation according to ROUGE
Natalie Schluter
In Proc of EACL, Valencia, Spain. pdf.

How (not) to train a dependency parser: The curious case of jackknifing part-of-speech taggers
Željko Agić and Natalie Schluter
In Proc of ACL, Vancouver, B.C., Canada. pdf.

Empirically sampling Universal Dependencies
Natalie Schluter and Željko Agić
In Proc of UDW, pages 117-122, Gothenburg, Sweden. pdf

2016

Approximate unsupervised summary optimisation for selections of ROUGE
Natalie Schluter and Héctor Martínez Alonso
In Proc of TALN, Paris, France. pdf.

Multilingual Projection for Parsing Truly Low-Resource Languages
Željko Agić, Anders Johannsen, Barbara Plank, Héctor Martínez Alonso, Natalie Schluter and Anders Søgaard
In TACL, vol 4, pages 301-312. pdf

CoastalCPH at Semeval 16 Task 11: The importance of designing your neural networks right
Joachim Bingel, Natalie Schluter and Héctor Martínez Alonso
In Semeval (NAACL), San Diego, CA, USA. pdf.

2015

Restarting Automata with Auxiliary Symbols Restricted by Lookahead Size.
Natalie Schluter
In International Journal of Computer Mathematics, Volume 92, Issue 5, pages 908-938. pdf

Unsupervised extractive summarization via coverage maximization with syntactic and semantic concepts.
Natalie Schluter and Anders Søgaard
In Proc of ACL, Beijing, China. pdf.

The complexity of finding the maximum spanning DAG and other restrictions for DAG parsing of natural language.
Natalie Schluter
In Proc of *SEM, Denver, CO, USA. pdf.

Maximising spanning subtree scores for parsing tree approximations of semantic dependency digraphs.
Natalie Schluter
In Proc of IWPT, Bilbao, Spain. pdf.

A Critical Survey on Measuring Success in Rank-Based Keyword Assignment to Documents.
Natalie Schluter
In Proc of TALN, Caen, France. pdf.

Effects of Graph Generation for Unsupervised Non-Contextual Single Document Keyword Extraction.
Natalie Schluter
In Proc of TALN, Caen, France. pdf.

2014

Copenhagen-Malmö:  Tree Approximations of Semantic Parsing Problems.
Natalie Schluter, Anders Søgaard, Jakob Elming, Dirk Hovy, Barbara Plank, Héctor Martinez Alonso, Anders Johanssen and Sigrid Klerke
In Proceedings of Semeval 2014, Dublin, Ireland. pdf

Centrality Measures for Non-Contextual Graph-Based Unsupervised Single Document Keyword Extraction.
Natalie Schluter
In Proceedings of TALN 2014, Marseilles, France. pdf

On maximum spanning DAG algorithms for semantic DAG parsing.
Natalie Schluter
In Proceedings of SP14, Baltimore, MD, USA. pdf

2011

Restarting Automata with Auxiliary Symbols and Small Lookahead.
Natalie Schluter
In Proceedings of LATA 2011, Tarragona, Spain. pdf

2010

On Lookahead Hierarchies for Monotone and Deterministic Restarting Automata with Auxiliary Symbols (Extended Abstract).
Natalie Schluter
In Proceedings of DLT 2010, London, Ontario, Canada.
LNCS, 2010, Volume 6224, 440-441. pdf

2009

A shorter proof that palindromes are not a Church-Rosser language, with extensions to almost-confluent and preperfect Thue systems.
Colm Ó Dúnlaing and Natalie Schluter
In Theoretical Computer Science, Vol. 411, Issue 3, pages 677-690. pdf

Dependency Parsing Resources for French: Converting Acquired Lexical Functional Grammar F-Structure Annotations and Parsing F-Structures Directly.
Natalie Schluter and Josef van Genabith
In Proceedings of NODALIDA 2009, Odense, Denmark. pdf

2008

Automatic Induction of Probabilistic LFG Resources for French.
Natalie Schluter and Josef van Genabith
In Proceedings of LREC08, Marrakesh, Morocco. pdf

2007

Preparing, Restructuring, and Augmenting a French Treebank: Lexicalised Parsers or Coherent Treebanks?
Natalie Schluter and Josef van Genabith
In Proceedings of PACLING 2007, Melbourne, Australia, pages 200–209. pdf

Theses

2011

Treebank-Based Deep Grammar Acquisition for French Probabilistic Parsing Resources.
Natalie Schluter
Thesis for the title of Doctor of Philosophy in Computer Applications
School of Computing, Dublin City University

2009

Restarting Automata, Context-free Languages, and Congruential Languages.
Natalie Schluter
Thesis for the title of Master of Science in Mathematics
School of Mathematics, Trinity College, University of Dublin

2005

The Role of Aspect in Paraphrase Operations
Natalie Schluter
Thesis for the title of Master of Arts in Linguistics
Département de linguistique et traduction, Université de Montréal