Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Attend this 59-minute Rothschild Public Lecture delivered by Professor Subhashis Ghoshal from North Carolina State University, exploring the intersection of Bayesian and frequentist statistical approaches through the concept of immersion posteriors. Learn how to achieve frequentist statistical goals while working within structural restrictions, a critical challenge in modern statistical inference. Discover advanced techniques for representing, calibrating, and leveraging prediction uncertainty that bridge traditional statistical methods with contemporary machine learning applications. Examine theoretical foundations and practical implementations of immersion posterior methods that maintain frequentist properties while accommodating complex structural constraints. Gain insights into cutting-edge research in statistical methodology that addresses fundamental questions about uncertainty quantification and prediction accuracy. Explore applications across various domains where structural restrictions naturally arise, from high-dimensional data analysis to machine learning model validation. This lecture is part of the Isaac Newton Institute's research programme on representing, calibrating, and leveraging prediction uncertainty from statistics to machine learning, bringing together leading researchers to advance understanding of uncertainty quantification in modern data science.
