Speaker: Bharat N. Anand, Vice Provost for Advances in Learning at Harvard University and the Henry R. Byers Professor of Business Administration at Harvard Business School
Title: “The Future of Education: Reflections from the Front Lines”
Abstract: In this talk I will offer a perspective on how, and why, digital technologies have shaped pedagogy and the practice of teaching over the last decade at Harvard, and what this means for the future. Specifically, I will examine: Where...
Please join us for our upcoming Statistics Seminar on May 19th with Art B. Owen who is a Max H. Stein Professor of Statistics at Stanford University.
Title: Tie-Breaker Designs
Abstract: Companies may offer incentives to their best customers and philanthropists may offer scholarships to the strongest students. They can evaluate the impact of these treatments later using a regression discontinuity analysis. Unfortunately, regression discontinuity analyses have high variance....
Abstract: Model-X knockoffs is a wrapper that transforms essentially any feature importance measure into a variable selection algorithm, which discovers true effects while rigorously controlling the expected fraction of false positives. A frequently discussed challenge to apply this method is to construct knockoff variables, which are synthetic variables obeying a crucial exchangeability property with the explanatory variables under study. This paper introduces techniques for knockoff generation in great...
Title: Recent developments on unbiased Monte Carlo methods
Abstract: Monte Carlo estimators, based on Markov chains or interacting particle systems, are typically biased when run with a finite number of iterations (or a finite number of particles). Although this is usually considered unavoidable, and negligible in the usual asymptotic sense, it is an important obstacle on the path towards scalable numerical integration on large-scale distributed computing systems. In a series of works that build on the seminal paper of Glynn and Rhee (2014), a...
Many applications, such as photon-limited imaging and genomics, involve large datasets with entries from exponential family distributions. It is of interest to estimate the covariance structure and principal components of the noiseless distribution. Principal Component Analysis (PCA), the standard method for this setting, can be inefficient for non-Gaussian noise. In this talk we present ePCA, a methodology for PCA on exponential family distributions. ePCA involves the eigendecomposition of a new covariance matrix estimator, constructed in a...
