I will give a high level talk about current research that studies how to prioritize cyber-security mitigations. I will talk about information technology supply chains, attack graphs, set cover variations, and budgeted maximization of submodular functions.
In a typical atmospheric contamination, “gas leaks” for instance, people are able to identify a deliberate pungent odor and the system responsible for the contamination contents have other various safe guards in place to protect environmental contamination. What happens when the contamination is deliberate and none of the safe guards …
Suppose G is a graph on N vertices, and H is a set of vertices on which someone has secretly “planted” a complete graph, or a graph whose edge-density is significantly greater than that of G. Can we efficiently and reliably find H, given G? This is called the “planted …
Understanding the way infections propagate through cells can allow us to know the severity of a disease and inform methods of treatment. The Human Rhinovirus (HRV) in particular is one virus which has been observed to spread in somewhat unusual ways which make it more difficult to characterize the rate …
A good code has both rate k/n and relative minimum distance d/n large. The fundamental question of coding theory is to describe the closure of the set of points (d/n,k/n). Goppa’s conjecture says that, except for isolated points, it’s the region below the curve y=1-H(x) for x
I will first explain to you the standard methodology for finding globally optimal solutions to nonconvex quadratic programs. I then will tell you a couple simple tricks, inspired by integer programming, that can speed up these algorithms by multiple orders of magnitude. If time allows, I can discuss a result …
Traffic stops are the most common contact that people will have with police. Most traffic stops are routine and end in either a warning or some form of citation for a traffic violation. However, a significant percentage of criminal charges arise out of what begins as a routine traffic stop. …
Sparse linear inverse problems appear in a variety of settings, but often the noise contaminating observations cannot accurately be described as bounded or arising from a Gaussian distribution. Poisson observations in particular are a characteristic feature of several real-world applications. Previous work on sparse Poisson inverse problems encountered several limiting …
I will discuss the influence of the seed in models of randomly growing trees; in particular, I will focus on the preferential attachment and uniform attachment models. In both of these models, perhaps surprisingly, different seeds lead to different distributions of limiting trees from a total variation point of view. …
Stochastic gradient descent algorithms for training linear and kernel predictors are gaining more and more importance, thanks to their scalability. While various methods have been proposed to speed up their convergence, the issue of the model selection phase has often been ignored in the literature. In fact, in theoretical works …