Google, IBM & Meta Certificates — All 10,000+ Courses at 40% Off
One annual plan covers every course and certificate on Coursera. 40% off for a limited time.
Get Full Access
Explore a lecture on sparsifying sums of norms presented by Yang Liu from Lawrence Berkeley National Laboratory as part of the "Beyond the Boolean Cube" series at the Simons Institute. Delve into the mathematical concept of creating a sparsified norm from a sum of multiple norms while maintaining a controlled error bound. Learn about the efficient algorithm for finding non-negative weights with high probability, and its implications for sparsifying sums of symmetric submodular functions. Discover how this approach extends to sparsifying sums of p-th powers of norms under certain smoothness conditions. Gain insights into the theoretical foundations and practical applications of norm sparsification in high-dimensional spaces.
