Hi, I'm Matt!
As of May 2020, I am a researcher at Two Sigma. Before that, I completed my Ph.D. at MIT in the EECS Department, where I was advised by Professor Stefanie Jegelka. My Ph.D. research focused on optimization (convex, submodular, robust, nonconvex, etc.) for machine learning.
I spent summer 2018 at Google NY working with Sashank Reddi, Satyen Kale, and Sanjiv Kumar. Before coming to MIT, I was at Stanford, where I graduated in 2015 with an M.S. in Electrical Engineering and a B.S. in Mathematics. I also spent summer 2014 at Microsoft Research Asia in Beijing, working with Thomas Moscibroda and Nic Lane.
Research
-
Edward Kim, Zach Jensen, Alexander van Grootel, Kevin Huang, Matthew Staib, Sheshera Mysore, Haw-Shiuan Chang, Emma Strubell, Andrew McCallum, Stefanie Jegelka, Elsa Olivetti.
Inorganic Materials Synthesis Planning with Literature-Trained Neural Networks. Journal of Chemical Information and Modeling, 2020.
.bib journal version arXiv@article{kim2020inorganic, author = {Kim, Edward and Jensen, Zach and Grootel, Alexander van and Huang, Kevin and Staib, Matthew and Mysore, Sheshera and Chang, Haw-Shiuan and Strubell, Emma and McCallum, Andrew and Jegelka, Stefanie and Olivetti, Elsa}, title = {Inorganic Materials Synthesis Planning with Literature-Trained Neural Networks}, journal = {Journal of Chemical Information and Modeling}, volume = {0}, number = {ja}, pages = {null}, year = {0}, doi = {10.1021/acs.jcim.9b00995}, note ={PMID: 31909619}, URL = {https://doi.org/10.1021/acs.jcim.9b00995}, eprint = {https://doi.org/10.1021/acs.jcim.9b00995} }
-
Matthew Staib, Stefanie Jegelka.
Distributionally Robust Optimization and Generalization in Kernel Methods. NeurIPS, 2019.
.bib NeurIPS arXiv code@incollection{staib2019mmddro, title = {Distributionally Robust Optimization and Generalization in Kernel Methods}, author = {Staib, Matthew and Jegelka, Stefanie}, booktitle = {Advances in Neural Information Processing Systems 32}, editor = {H. Wallach and H. Larochelle and A. Beygelzimer and F. d\textquotesingle Alch\'{e}-Buc and E. Fox and R. Garnett}, pages = {9131--9141}, year = {2019}, publisher = {Curran Associates, Inc.}, url = {http://papers.nips.cc/paper/9113-distributionally-robust-optimization-and-generalization-in-kernel-methods.pdf}}
-
Matthew Staib, Sashank J. Reddi, Satyen Kale, Sanjiv Kumar, Suvrit Sra.
Escaping Saddle Points with Adaptive Gradient Methods. ICML, 2019.
.bib PMLR arXiv@inproceedings{staib2019escaping, title = {Escaping Saddle Points with Adaptive Gradient Methods}, author = {Staib, Matthew and Reddi, Sashank and Kale, Satyen and Kumar, Sanjiv and Sra, Suvrit}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {5956--5965}, year = {2019}, editor = {Chaudhuri, Kamalika and Salakhutdinov, Ruslan}, volume = {97}, series = {Proceedings of Machine Learning Research}, address = {Long Beach, California, USA}, month = {09--15 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v97/staib19a/staib19a.pdf}, url = {http://proceedings.mlr.press/v97/staib19a.html} }
-
Matthew Staib, Stefanie Jegelka.
Robust Budget Allocation via Continuous Submodular Functions. Applied Mathematics & Optimization, Special Issue on Optimization for Data Sciences, 2019.
.bib journal version@Article{staib2019robust, author="Staib, Matthew and Jegelka, Stefanie", title="Robust Budget Allocation Via Continuous Submodular Functions", journal="Applied Mathematics {\&} Optimization", year="2019", month="Mar", day="29", issn="1432-0606", doi="10.1007/s00245-019-09567-0", url="https://doi.org/10.1007/s00245-019-09567-0" }
-
Matthew Staib*, Bryan Wilder*, Stefanie Jegelka.
Distributionally Robust Submodular Maximization. AISTATS, 2019.
.bib PMLR arXiv@inproceedings{staib2018distributionally, author = {Staib, Matthew and Wilder, Bryan and Jegelka, Stefanie}, booktitle = {Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics}, title = {Distributionally Robust Submodular Maximization}, volume = {89}, series = {Proceedings of Machine Learning Research}, year = {2019} }
-
Matthew Staib, Sebastian Claici, Justin Solomon, Stefanie Jegelka.
Parallel Streaming Wasserstein Barycenters. NIPS, 2017.
.bib NIPS arXiv code@incollection{staib2017parallel, title = {Parallel Streaming Wasserstein Barycenters}, author = {Staib, Matthew and Claici, Sebastian and Solomon, Justin M and Jegelka, Stefanie}, booktitle = {Advances in Neural Information Processing Systems 30}, editor = {I. Guyon and U. V. Luxburg and S. Bengio and H. Wallach and R. Fergus and S. Vishwanathan and R. Garnett}, pages = {2644--2655}, year = {2017}, publisher = {Curran Associates, Inc.}, url = {http://papers.nips.cc/paper/6858-parallel-streaming-wasserstein-barycenters.pdf} }
-
Matthew Staib, Stefanie Jegelka.
Robust Budget Allocation via Continuous Submodular Functions. ICML, 2017. (Superseded by journal version)
.bib PMLR arXiv code@inproceedings{staib2017robust, title = {Robust Budget Allocation via Continuous Submodular Functions}, author = {Matthew Staib and Stefanie Jegelka}, booktitle = {Proceedings of the 34th International Conference on Machine Learning}, pages = {3230--3240}, year = {2017}, editor = {Doina Precup and Yee Whye Teh}, volume = {70}, series = {Proceedings of Machine Learning Research}, address = {International Convention Centre, Sydney, Australia}, month = {06--11 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v70/staib17a/staib17a.pdf}, url = {http://proceedings.mlr.press/v70/staib17a.html} }
-
Tianlin Shi, Forest Agostinelli, Matthew Staib, David Wipf, Thomas Moscibroda.
Improving Survey Aggregation with Sparsely Represented Signals. KDD, 2016.
.bib .pdf@inproceedings{Shi:2016:ISA:2939672.2939876, author = {Shi, Tianlin and Agostinelli, Forest and Staib, Matthew and Wipf, David and Moscibroda, Thomas}, title = {Improving Survey Aggregation with Sparsely Represented Signals}, booktitle = {Proceedings of the 22Nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining}, series = {KDD '16}, year = {2016}, isbn = {978-1-4503-4232-2}, location = {San Francisco, California, USA}, pages = {1845--1854}, numpages = {10}, url = {http://doi.acm.org/10.1145/2939672.2939876}, doi = {10.1145/2939672.2939876}, acmid = {2939876}, publisher = {ACM}, address = {New York, NY, USA}, keywords = {compressive sensing, deep learning, james stein estimator, multi-task learning, presidential elections, survey aggregation}, }
-
Matthew Staib*, Bryan Wilder*, Stefanie Jegelka.
Distributionally Robust Submodular Maximization. ICML Workshop on Modern Trends in Nonconvex Optimization for Machine Learning, 2018. Spotlight.
.bib arXiv@inproceedings{staib2018distributionally, author = {Staib, Matthew and Wilder, Bryan and Jegelka, Stefanie}, title = {Distributionally Robust Submodular Maximization}, booktitle = {ICML 2018 Workshop on Modern Trends in Nonconvex Optimization for Machine Learning}, year = {2018} }
-
Matthew Staib, Stefanie Jegelka.
Distributionally Robust Deep Learning as a Generalization of Adversarial Training. NIPS Machine Learning and Computer Security Workshop, 2017.
.bib@inproceedings{staib2017distributionally, author = {Staib, Matthew and Jegelka, Stefanie}, title = {Distributionally Robust Deep Learning as a Generalization of Adversarial Training}, booktitle = {NIPS Machine Learning and Computer Security Workshop}, year = {2017} }
-
Matthew Staib, Stefanie Jegelka.
Wasserstein k-means++ for Cloud Regime Histogram Clustering. Climate Informatics, 2017.
.bib .pdf code@inproceedings{staib2017wasserstein, author = {Staib, Matthew and Jegelka, Stefanie}, title = {Wasserstein k-means++ for Cloud Regime Histogram Clustering}, booktitle = {Proceedings of the Seventh International Workshop on Climate Informatics: CI 2017}, year = {2017} }
Conference and Journal Papers
Workshops
* indicates equal contribution.
Contact
mstaib [at] hey [dot] com