All are invited to attend this ILE event featuring a luncheon and oral presentations by students.
11:55-12:05 Gathering and lunch begins
12:05-12:15 Michael Dunlea- Optimal Control Theory
12:18-12:28 Jon Mellinger- Statistical search of the MVP in the NBA
12:30-12:40 Alexandra Signoriello- Optimization Techniques with Applications to Gene Interactions and Nosocomial Infection
12:43-12:53 Magdalena Parks- Density and Fault Tolerance in Sensor Networks
Optimal Control Theory was developed to provide optimization methods for a dynamic system, with control functions and under certain constraints, in order to achieve and optimize a certain output. This presentation will walk through the steps of Optimal Control Theory such as state variables, control variable, state equation, objective function, transversality conditions, Hamilitonian, optimality condition, Pontryagin's maximum principle, Jacobi Matrix. Examples include application to VRE infections in hospital intensive care units.
With my internship and SAS I am looking at an up to date MVP (Most Valuable Player) model through November (the first month of the regular season) for the NBA (National Basketball Association), using all of the statistics that we as interns gather from every game. I will compare it to a different MVP model and say how improvements could be made upon my model and what I would change for any fine tuning. For example, a minute per game limit, which will be explained in the presentation.
Mathematical models are used to illustrate the behavior of complex dynamical systems. This project focuses on using optimization techniques to find parameters that best fit mathematical models to given experimental data. This makes the prediction of the models more accurate. Simulations search for parameters that minimize the residuals and to visualize the best fit. The simulations will be applied to two biological systems: Gene interactions in a regulatory response network, and transmission of a nosocomial infection.
A wireless sensor network is a collection of sensors that can gather information about the surrounding environment. These wireless sensor networks are given the task of collecting data and ensuring that the data is passed along through the network to get to its destination. The wireless sensors can unexpectedly fail, and if the network isn't prepared to deal with that failure, major problems may arise. Fault tolerance is the ability of a system to minimize the impact of these failures. In dense networks, each sensor is in contact with more sensors than in a sparse network. Dense wireless sensor network should be better able to recover from a sensor failure than a sparse network.
Friday, Nov 30, 2012