Yasemin Altun
Research Assistant Professor
altun@tti-c.org

Toyota Technological Institute at Chicago
University of Chicago Press Bldg
1427 East 60th Street
Chicago IL, 60637


Short Bio  :       I received a Master's Degree in Cognitive Science from Middle East Technical University in Ankara, Turkey in 1999 and a Master's Degree in Computer Science from Brown University in 2003. I completed my graduate work with Thomas Hofmann to receive my Ph.D. in Machine Learning from Brown University in 2005. Currently, I am a Research Assistant Professor of Computer Science at the Toyota Technological Institute (more details).

Research Interests  :       My current research focuses on Machine Learning, in particular supervised and semi-supervised structured learning. But my interests also cover related areas like Information Theory, Information Retrieval, Natural Language Processing and Computational Biology.


Teaching

CMCS 35900 Topics in AI - Generative and Discriminative Approaches for Graphical Models

Publications

2006

Unifying Divergence Minimization and Statistical Inference via Convex Duality
Yasemin Altun and Alex Smola
in the 19th Annual Conference on Learning Theory, 2006.

Broad-Coverage Sense Disambiguation and Information Extraction with a Supersense Sequence Tagger
Massimiliano Ciaramita and Yasemin Altun
in Empirical Methods in Natural Language Processing (EMNLP), 2006

Transductive Gaussian Process Regression with Automatic Model Selection
Quoc Le, Alex Smola, Thomas Gaertner, Yasemin Altun
in European Conference on Machine Learning(ECML), 2006. Best paper award.

Estimating Conditional Densities of Structured Outputs in RKHS
Yasemin Altun and Alex Smola
in G. Bakir, T. Hofmann, B. Scholkopf, A.J. Smola, B. Taskar, and S.V.N. Vishwanathan, editors,
Machine Learning with Structured Outputs, MIT Press, 2006.

Large Margin Methods for Structured and Interdependent Output Variables,
Yasemin Altun and Thomas Hofmann and Ioannis Tsochantaridis
in G. Bakir, T. Hofmann, B. Scholkopf, A.J. Smola, B. Taskar, and S.V.N. Vishwanathan, editors,
Machine Learning with Structured Outputs, MIT Press, 2006.

Unifying Divergence Minimization and Statistical Inference via Convex Duality
Yasemin Altun and Markus Hegland and Alex Smola
in preparation for IEEE Transactions on Information Theory.


2005

Maximum Margin Semi-Supervised Learning for Structured Variables
Yasemin Altun, David McAllester and Mikhail Belkin
Advances in Neural Information Processing Systems (NIPS*17), 2005

Named-Entity Recognition in Novel Domains with External Lexical Knowledge
Massimiliano Ciaramita and Yasemin Altun
Advances in Structured Learning for Text and Speech Processing Workshop,
Advances in Neural Information Processing Systems (NIPS*17), 2005

Semi-Supervised Structure Learning
Yasemin Altun and David McAllester
Learning at Snowbird, 2005

Large Margin Methods for Structured and Interdependent Output Variables,
Ioannis Tsochantaridis, Thorsten Joachims, Thomas Hofmann & Yasemin Altun
Journal of Machine Learning Research (JMLR), 6(Sep):1453-1484, 2005.

Discriminative Methods for Label Sequence Learning
Ph.D. thesis from Brown University, Department of Computer Science.

2004

Gaussian Process Classification for Segmenting and Annotating Sequences
Yasemin Altun, Thomas Hofmann & Alex Smola
21th International Conference on Machine Learning (ICML), 2004
Here is an earlier version of this paper as a techical report with matrix operation represented termwise.

Exponential Families for Conditional Random Fields
Yasemin Altun, Alex Smola & Thomas Hofmann
20th Conference on Uncertainty in Artificial Intelligence (UAI),  2004

Using Conditional Random Fields to Predict Pitch Accent in Conversational Speech
Michelle Gregory & Yasemin Altun
42nd Annual Meeting of the Association for Computational Linguistics (ACL), 2004

Support Vector Machine Learning for Interdependent and Structured Output Spaces
Ioannis Tsochantaridis, Thomas Hofmann, Thorsten Joachims & Yasemin Altun
21th International Conference on Machine Learning (ICML), 2004

2003

Loss Functions and Optimization Methods for Discriminative Learning of Label Sequences
Yasemin Altun, Mark Johnson & Thomas Hofmann
Empirical Methods in Natural Language Processing (EMNLP), 2003

Hidden Markov Support Vector Machines
Yasemin Altun, Ioannis Tsochantaridis & Thomas Hofmann
20th International Conference on Machine Learning (ICML), 2003

Large Margin Methods for Label Sequence Learning
Yasemin Altun & Thomas Hofmann
8th European Conference on Speech Communication and Technology (EuroSpeech), 2003

2002


Discriminative Learning for Label Sequences via Boosting

Yasemin Altun, Thomas Hofmann & Mark Johnson
Advances in Neural Information Processing Systems (NIPS*15), 2003

A crossover between SVMs and HMMs for protein structure prediction
Ioannis Tsochantaridis, Yasemin Altun & Thomas Hoffman
Advances in Neural Information Processing Systems, Bioinformatics Workshop, 2002.

Learning over Discrete Output Spaces via Joint Kernel Functions
Thomas Hofmann, Ioannis Tsochantaridis and Yasemin Altun
Advances in Neural Information Processing Systems, Workshop on Kernel Methods, 2002.

2001

Inducing SFA with Epsilon-Translations Using Minimum Description Length
Yasemin Altun, Mark Johnson
Finite State Methods in Natural Language Processing Workshop, ESSLLI 2001.


2000

Reading Comprehension Programs in a Statistical-Language-Processing Class
E. Charniak, Y. Altun, R. de Salvo Braz,
B. Garrett, M. Kosmala, T. Moscovich, L. Pang, C. Pyo, Y. Sun, W. Wy, Z. Yang, S. Zeller, and L. Zorn.
ANLP-NAACL Workshop on Reading Comprehension Tests as Evaluation for Computer-Based Language Understanding Systems, 2000.