September 13, 2016 to September 14, 2016
Speaker: Suvrit Sra (MIT)
In this talk, I will highlight some aspects of geometry and its role in optimization. In particular, I will talk about optimization problems whose parameters are constrained to lie on a manifold or in a suitable metric space. These...
September 27, 2016 to September 28, 2016
Speaker: Steve Wright (University of Wisconsin-Madison)
A permutation of n components appears among the variables or in the formulation of several interesting optimization problems, including quadratic assignment, 2-SUM, and projection onto the unit simplex or an l-1 ball. One device used to...
October 4, 2016 to October 5, 2016
Speaker: Frank Kschischang (University of Toronto)
The vast majority of the world's telecommunications and Internet traffic is carried, for at least part of its journey, over a network of land-based and under-sea fiber-optic cables that span the globe. Recent decades have witnessed steady...
October 25, 2016 to October 26, 2016
Speaker: Jeffrey Bilmes (University of Washington)
In this talk, we'll first review how submodular functions are useful in data science for various data manipulation problems (e.g., summarization and partitioning), and how certain submodular functions (e.g., sums of concave composed with...
November 1, 2016 to November 2, 2016
Speaker: Bruce Hajek (University of Illinois Urbana-Champaign)
Detecting or estimating a dense community from a network graph offers a rich set of problems involving the interplay of algorithms, complexity, and information limits. The speaker in his talk will present an overview and recent results on...
November 15, 2016 to November 16, 2016
Speaker: Negar Kiyavash (University of Illinois Urbana-Champaign)
Learning the influence structure of multiple time series data is of great interest to many disciplines. We discuss approaches for learning causal interaction network of mutually exciting Hawkes processes. In such processes, the arrival of an...
November 22, 2016 to November 23, 2016
Speaker: Patrick Rebeschini (Yale)
The complexity of network optimization depends on the network topology, the nature of the objective function, and what information (local or global) is available to the decision makers. In this talk we introduce a notion of network locality...
November 29, 2016 to November 30, 2016
Speaker: Rebecca Willett (University of Wisconsin-Madison)
Vector autoregressive models characterize a variety of time series in which linear combinations of current and past observations can be used to accurately predict future observations. For instance, each element of an observation vector could...
