Speaker 1: Yun-Ju Cheng

Title

The use of propensity scores in assessing generalizability of RCTs to a target population

Abstract

While properly planned and conducted randomized trials yield accurate estimate of the treatment effect for the trial samples, it remains susceptible to a lack of external validity. An inverse-propensity-score-weighting approach was proposed by Cole and Stuart (2010) to generalize inferences from a randomized clinical trial to a specified target population, on the aspect of generalizability that is related to differences in the characteristics of trial samples and target population.

Speaker 2: Ahmad Mousavi

Title

An introduction to Compressive Sensing

Abstract

In this talk, we explain about the idea of compressive sensing which is built on the assumption of sparsity. Although this original problem is NP-hard, for some good matrices it is possible to solve it via the linear programming or other suitable algorithms. After giving some tractable algorithms to solve the original problem, we shortly talk about the mutual coherence of the measurement matrix; a tool to guarantee their convergence. Then, we have some theorems about the convergence of the given algorithms by means of mutual coherence.