Systems | Information | Learning | Optimization

Linear Dueling Bandits

Dueling bandits is a variant of the multi-armed bandit problem where instead of playing an arm and observing the reward at each instant, you duel two arms (pairwise comparison) and observe the winner among the two. Linear bandits is a special case of contextual bandits where the reward of an …

Urban Areas LiDar Scanning

I will include the following points in my talk: – What is LiDar? – LiDar Types. – Special considerations for urban areas aerial scanning. – Applications of aerial scanning for urban areas.

Multi-Resolution Spatio-Temporal Analysis of Atmospheric Contamination

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 …

Convex relaxations and the planted clique problem

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 …

Modeling the Spread of Human Rhinovirus

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 …

The fundamental question of coding theory

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

GuBoLi — The World’s Greatest Solver for Box-Constrained Nonconvex Quadratic Programs

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 …

Law 101: What to Do When Pulled Over by Police

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. …

A data-dependent weighted LASSO

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 …