We present a simple technique for estimating parts of the Singular Value Decomposition (SVD) of a data matrix from a small randomly compressed “sketch” of that matrix. In sensor network settings–where each column of the data matrix comes from a separate sensor–the sketch can be assembled using operations local to …
Large-scale computational simulation using numerical methods is now widespread in science and engineering. Good approximation methods exist for many problems and fast linear solvers for the resulting linear(ized) equations too. In this talk we will focus on finite element approximations and iterative linear equation solvers. We intend to show how …
An extremely powerful tool in the field of topological data analysis (TDA) is persistent homology which seeks to quantify structure in a large data set. In particular, persistence can find homology classes for an underlying manifold even when obscured by noise. These homology classes can then give information about the …
Many scientific and engineering applications rely upon the accurate reconstruction of spatially, spectrally, and temporally distributed phenomena from photon-limited data. When the number of observed events is very small, accurately extracting knowledge from this data requires the development of both new computational methods and novel theoretical analysis frameworks. This task …
In various decision making problems, one is given the advice or predictions of several classifiers of unknown reliability, over multiple questions or queries. This scenario is different from the standard supervised setting where classifier accuracy can be assessed using available labeled training or validation data, and raises several questions: given …
We consider the problem of learning a directed acyclic graph (DAG) model based on conditional independence testing. The most widely used approaches to this problem are variants of the PC algorithm. One of the drawbacks of the PC algorithm is that it requires the strong-faithfulness assumption, which has been show …
What if there is a teacher who knows the learning goal and wants to design good training data for a machine learner? In this talk, I formalize the optimal teaching problem. In particular, I will focus on learners who employ Bayesian models. The framework is expressed as an optimization problem …
We study decentralized markets which include middlemen, producers and consumers connected via a trading network. We develop a model for trade in such settings based on non-cooperative bargaining with search frictions. Our goal is to investigate how the structure of the trading network and the role of middlemen influence the …
A wide-ranging and freewheeling panel discussion of how science, especially the science of information and optimization, can best make itself known to the public.
Speaker 1: Chris Hinrichs Abstract: Multiple hypothesis testing is a significant problem in nearly all neuroimaging studies. In order to correct for this phenomena, we require a reliable estimate of the Family-Wise Error Rate (FWER). The well known Bonferroni correction method, while being simple to implement, is quite conservative, and …