Introduction to Algorithms For Big Data Compsci 229r Lecture 12
Let's dive into the details surrounding Algorithms For Big Data Compsci 229r Lecture 12. Alon's JL lower bound, beyond worst case analysis: suprema of gaussian processes, Gordon's theorem.
Algorithms For Big Data Compsci 229r Lecture 12 Comprehensive Overview
Competitive paging, cache-oblivious ORS theorem (distributional JL implies Gordon's theorem), sparse JL. Khintchine, decoupling, Hanson-Wright, proof of distributional JL lemma.
Distinct elements, k-wise independence, geometric subsampling of streams.
Summary & Highlights for Algorithms For Big Data Compsci 229r Lecture 12
- FPTAS (knapsack), FPRAS (DNF counting), semidefinite programming, Goemans-Williamson MAXCUT
- Approximate matrix multiplication with Frobenius error via sampling / JL, matrix median trick, subspace embeddings.
- Necessity of randomized/approximate guarantees, linear sketching, AMS sketch, p-stable sketch for p less than 2.
- External memory model: linked list, matrix multiplication, B-tree, buffered repository tree, sorting.
- Linear least squares via subspace embeddings, leverage score sampling, non-commutative Khintchine, oblivious subspace ...
That wraps up our extensive overview of Algorithms For Big Data Compsci 229r Lecture 12.
