Consortium for Modeling and Analysis of Treatments and Interventions

Lev Reyzin is an Assistant Professor in the MCS group at UIC's mathematics department. His research focuses on computational and statistical learning, but he is more broadly interested in topics ranging from practical issues in machine learning to the theoretical foundations of computer science. Previously, Lev was a Simons Postdoctoral Fellow at Georgia Tech, and before that, an NSF CI-Fellow at Yahoo! Research, where he tackled problems in computational advertising. Lev received his Ph.D. from Yale under Dana Angluin and his undergraduate degree from Princeton. His work has earned awards at ICML, COLT, and AISTATS, as well as several national fellowships.

UNDER CONSTRUCTION**

Spring 2014: Emory MathCS. "Weights and Measures: Prediction in the Era of Big Data."
Spring 2014: ITA. "Training-Time Optimization of a Budgeted Booster."
Fall 2013: ISAIM. "On Boosting Sparse Parities."
Fall 2013: MSR-NYC. "On the Resilience of Biparite Networks."
Fall 2013: UIC Math. "On Finding Planted Cliques and Solving Random Linear Equations."
Spring 2013: CASSC Plenery Talk. "Bandit Algorithms for Internet-Scale Applications."
Spring 2013: TTI. "New Algorithms for Contextual Bandits."
Summer 2012: Stony Brook, MSR-NYC. "Statistical Algorithms and the Planted Clique Problem"
Spring/Summer/Fall 2012: CMU, Google Research, UAlberta, UIC. "New Algorithms for Contextual Bandits."
Spring 2012: UIC MSCS. "The Complexity of Statistical Algorithms."
Spring 2012: Sandia Labs, Bell Labs, William & Mary, MIT-LL. "From Queries to Bandits: Learning by Interacting."
Fall 2011: ALT. "On Noise-Tolerant Learning of Sparse Parities and Related Problems."
Fall 2011: UPenn, Yale, Georgia Tech. "The Complexity of Statistical Algorithms." (updated)
Summer 2011: ICML. "Boosting on a Budget: Sampling for Feature-Efficient Prediction."
Spring 2011: AISTATS. "Contextual Bandit Algorithms with Supervised Learning Guarantees."
Fall 2010: ALT. "Lower Bounds on Learning Random Structures with Statistical Queries."
Fall 2010: ALT. "Inferring Social Networks from Outbreaks."
Summer 2010: Ben Gurion Univ., Yahoo! Research, Georgia Tech. "New Algorithms for Contextual Bandits."
Summer 2010: ICML/COLT Budgeted Learning Workshop. "Boosting on a Feature Budget."
Spring 2010: ARC at Georgia Tech. "Active Learning of Interaction Networks."
Spring 2010: Santa Fe Institute. "Learning Social Networks, Actively and Passively."
Spring 2010: IBM TJ Watson. "Learning Analog Circuits, Graphical Models, and Social Networks by Injecting Values."
Fall 2009: ALT. "Learning Finite Automata Using Label Queries."
Summer 2009: Thesis Defense. "Active Learning of Interaction Networks."
Fall 2008: ALT. "Optimally Learning Social Networks with Activations and Suppressions."
Summer 2008: COLT. "Learning Acyclic Probabilistic Circuits Using Test Paths."
Spring 2008: Yahoo! Research NY. "Learning Hidden Circuits and (Social) Networks by Injecting Values."
Fall 2007: Machine Learning Lunch at UMass Amherst. "Learning Hidden Graphs and Circuits with Query Access."
Summer 2007: COLT. "Learning Large-Alphabet and Analog Circuits with Value Injection Queries."
Fall 2006: Yale. "Learning Graphs with Queries."
Summer/Fall 2006: ICML, Princeton Univ., NYAS, Yale. "How Boosting the Margin Can Also Boost Classifier Complexity."