Our research focuses on statistical inference and estimation, principled approximation methods for
problems with limited computational resources, analysis and design of
algorithms for various modern estimation problems such as those
involving predominantly incomplete data.
On the applied side, our work involves primarily problems in natural
language processing, computational biology (e.g., regulatory models),
recommender and other large scale inference problems, as well as information
retrieval.

__People (students/postdocs*)__

David Alvarez,
Andreea Gane,
Vikas Garg,
Paresh Malalur,
Jonas Mueller,
David Reshef,

D. Alvarez-Melis and T. Jaakkola.

**Tree structured decoding with doubly recurrent neural networks. **

In * International Conference on Learning Representations (ICLR)*, 2017.

[pdf]
J. Mueller, D. Reshef, G. Du, and T. Jaakkola.

**Learning optimal interventions. **

In * Artificial Intelligence and Statistics (AISTATS)*, 2017.

[pdf]
V. Garg and T. Jaakkola.

**Learning tree structured potential games. **

In * Advances in Neural Information Processing Systems (NIPS)*, 2016.

[pdf]
T. Lei, R. Barzilay, and T. Jaakkola.

**Rationalizing neural predictions. **

In * Empirical Methods in Natural Language Processing (EMNLP)*, 2016.

[pdf]
Y. Gu, R. Barzilay, and T. Jaakkola.

**Food adulteration detection using neural networks. **

In * Empirical Methods in Natural Language Processing (EMNLP)*, 2016.

J. Honorio and T. Jaakkola.

**Structured prediction: From gaussian perturbations to linear-time principled algorithms. **

In * Uncertainty in Artificial Intelligence (UAI)*, 2016.

[pdf]
T. Hashimoto, D. Alvarez-Melis, and T. Jaakkola.

**Word embeddings as metric recovery in semantic spaces. **

* Transactions of the Association for Computational Linguistics (TACL)*, 4, 2016.

[pdf]
T. Hashimoto, T. Jaakkola, and D. Gifford.

**Learning population-level diffusions with generative {RNN}s. **

In * International Conference on Machine Learning (ICML)*, 2016.

[pdf]
Y. Zhang, D. Gaddy, R. Barzilay, and T. Jaakkola.

**Ten pairs to tag -- multilingual pos tagging via coarse mapping between embeddings. **

In * The 15th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL)*, 2016.

[pdf]
T. Lei, H. Joshi, R. Barzilay, T. Jaakkola, K. Tymoshenko, A. Moschitti, and L. Marquez.

**Semi-supervised question retrieval with gated convolutions. **

In * The 15th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL)*, 2016.

[pdf]
V. Garg, C. Rudin, and T. Jaakkola.

**Craft: Cluster-specific assorted feature selection. **

In * Artificial Intelligence and Statistics (AISTATS)*, 2016.

[pdf]
T. Hashimoto, D. Alvarez-Melis, and T. Jaakkola.

**Word, graph and manifold embedding from markov processes. **

In * arXiv:1509.05808*, 2015.

[link]
T. Hashimoto, Y. Sun, and T. Jaakkola.

**From random walks to distances on unweighted graphs. **

In * Advances in Neural Information Processing Systems (NIPS)*, 2015.

[pdf]
J. Mueller and T. Jaakkola.

**Principal differences analysis: Interpretable characterization of differences between distributions. **

In * Advances in Neural Information Processing Systems (NIPS)*, 2015.

[pdf]
T. Lei, R. Barzilay, and T. Jaakkola.

**Molding {CNN}s for text: Non-linear, non-consecutive convolutions. **

In * Empirical Methods in Natural Language Processing*, 2015.

[pdf] [link]
K. Narasimhan, R. Barzilay, and T. Jaakkola.

**An unsupervised method for uncovering morphological chains. **

* Transactions of the Association for Computational Linguistics*, 3:157--167, 2015.

[pdf] [link]
T. Hashimoto, Y. Sun, and T. Jaakkola.

**Metric recovery from directed unweighted graphs. **

In * Artificial Intelligence and Statistics*, 2015.

[pdf]
Y. Xin and T. Jaakkola.

**Controlling privacy in recommender systems. **

In * Advances in Neural Information Processing Systems*, 2014.

[pdf]
Y. Zhang, T. Lei, R. Barzilay, and T. Jaakkola.

**Greed is good if randomized: New inference for dependency parsing. **

In * Conference on Empirical Methods in Natural Language Processing (EMNLP)*, 2014.

[pdf]
T. Lei, Y. Xin, Y. Zhang, R. Barzilay, and T. Jaakkola.

**Low-rank tensors for scoring dependency structures. **

In * Association for Computational Linguistics*, 2014.

[pdf]
Y. Zhang, T. Lei, R. Barzilay, T. Jaakkola, and A. Globerson.

**Steps to excellence: Simple inference with refined scoring of dependency trees. **

In * Association for Computational Linguistics*, 2014.

[pdf]
J. Honorio and T. Jaakkola.

**A unified framework for consistency of regularized loss minimizers. **

In * Proceedings of the 31th International Conference on Machine Learning*, 2014.

[pdf]
A. Gane, T. Hazan, and T. Jaakkola.

**Learning with maximum a-posteriori perturbation models. **

In * Artificial Intelligence and Statistics*, 2014.

[pdf]
S. Maji, T. Hazan, and T. Jaakkola.

**Active boundary annotation using random map perturbations. **

In * Artificial Intelligence and Statistics*, 2014.

[pdf]
J. Honorio and T. Jaakkola.

**Tight bounds for the expected risk of linear classifiers and pac-bayes finite-sample guarantees. **

In * Artificial Intelligence and Statistics*, 2014.

[pdf]
F. Orabona, T. Hazan, A. Sarwate, and T. Jaakkola.

**On measure concentration of random maximum a-posteriori perturbations. **

In * Proceedings of the 31th International Conference on Machine Learning*, 2014.

O. Meshi, T. Jaakkola, and A. Globerson.

**Smoothed coordinate descent for map inference. **

In S. Nowozin, P. V. Gehler, J. Jancsary, and C. Lampert, editors, * Advanced Structured Prediction*. MIT Press, 2014.

[pdf]
R. Sherwood, T. Hashimoto, C. O'Donnell, S. Lewis, A. Barkal, J.P. van Hoff, V. Karun, T. Jaakkola, and D. Gifford.

**Discovery of directional and nondirectional pioneer transcription factors by modeling dnase profile magnitude and shape. **

* Nature Biotechnology*, 32(2):171--178, 2014.

[pdf]
T. Hazan, S. Maji, J. Keshet, and T. Jaakkola.

**Learning efficient random maximum a-posteriori predictors with non-decomposable loss functions. **

In * Advances of Neural Information Processing Systems*, 2013.

[pdf]
T. Hazan, S. Maji, and T. Jaakkola.

**On sampling from the gibbs distribution with random maximum a posteriori perturbations. **

In * Advances of Neural Information Processing Systems*, 2013.

[pdf]
J. Honorio and T. Jaakkola.

**Two-sided exponential concentration bounds for bayes error rate and shannon entropy. **

In * Proceedings of the 30th International Conference on Machine Learning*, 2013.

[pdf]
J. Honorio and T. Jaakkola.

**Inverse covariance estimation for high-dimensional data in linear time and space: Spectral methods for riccati and sparse models. **

In * Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence*, 2013.

[pdf]
T. Hazan and T. Jaakkola.

**On the partition function and random maximum a-posteriori perturbations. **

In * Proceedings of the 29th International Conference on Machine Learning (ICML)*, 2012.

[pdf]
O. Meshi, T. Jaakkola, and A. Globerson.

**Convergence rate analysis of map coordinate minimization algorithms. **

In * Advances in Neural Information Processing Systems*, 2012.

T. Hashimoto, T. Jaakkola, R. Sherwood, E. Mazzoni, H. Witchterle, and D. Gifford.

**Lineage based identification of cellular states and expression programs. **

In * Proceedings of the 20th Annual International Conference on Intelligent Systems for Molecular Biology (ISMB)*, 2012.

[pdf]
Z. Kolter and T. Jaakkola.

**Approximate inference in additive factorial hmms with application to energy disaggregation. **

* Proceedings of the 15th International Conference on Artificial Intelligence and Statistics, JMLR WCP*, 22:1472--1482, 2012.

[pdf]
Y. Xin and T. Jaakkola.

**Primal-dual methods for sparse constrained matrix completion. **

* Proceedings of the 15th International Conference on Artificial Intelligence and Statistics, JMLR WCP*, 22:1323--1331, 2012.

[pdf]