AI, Data Science & Cloud Certificates from Google, IBM & Meta — 40% Off
One plan covers every Professional Certificate on Coursera. 40% off Coursera Plus Annual.
Unlock All Certificates
Explore a comprehensive lecture on Stein Variational Gradient Descent (SVGD) and its fast finite-particle convergence properties. Delve into advanced optimization techniques in data science as part of the "Data Science: Probabilistic and Optimization Methods" discussion meeting. Learn from expert Dheeraj Nagaraj of the International Centre for Theoretical Sciences as he presents cutting-edge research on SVGD, a powerful algorithm for approximate Bayesian inference. Gain insights into the theoretical foundations and practical applications of this method in machine learning and statistical inference. Understand how SVGD combines the strengths of variational inference and particle-based methods to achieve efficient and accurate posterior approximations. Discover the latest developments in convergence analysis for SVGD with a finite number of particles, and explore its implications for scalable Bayesian computation in high-dimensional settings.
