Ravi Kannan
where else to find them
what they talk about
where they keep going back
no series more than once — all 6 in the filters
on a stage with
who talks about the same things
nobody close enough yet
| title | conference | date |
|---|---|---|
| Latent Variable models and Subset Smoothing | The Role of TCS in Modern Machine Learning | 2026-05-26 |
Defines Subset Smoothed polytope K' as the convex hull of averages over k‑subsets of the n noisy samples and proves K' ≈ K when many latent points lie near… | ||
| Detecting Weak Signals via Hypothesis Testing | Robust and High-Dimensional Statistics (Simons Institute workshop, 2018) | 2018-10-29 |
Binary test between product noise P0 and a weak mixture signal (1−ε)P0+εP1 uses the chi‑square distance D = χ2(P1||P0) to set a sharp threshold: if ε√D → 0 no… | ||
| Intro and Foundations of Data Science I | Foundations of Data Science Boot Camp 2018 | 2018-08-27 |
A compact course first promotes a finite-moment method that replaces exponential moment proofs with Rth moments and an R! bound to get Gaussian-like moment… | ||
| Foundations of Data Science II | Foundations of Data Science Boot Camp 2018 | 2018-08-27 |
Claims that SVD identifies the k-dimensional subspace spanned by Gaussian means, yielding a polynomial-time algorithm to learn K spherical Gaussians with… | ||
| Topic Modeling: A Provable Spectral Method | Spectral Algorithms: From Theory to Practice 2014 | 2014-10-27 |
Provides a provable pipeline that recovers K topic vectors by combining threshold then SVD with k‑means and a catchword extraction step, under four explicit… | ||
| Optimizing with Tensors | Workshop on Succinct Data Representations and Applications (Simons Institute Big Data 2013) | 2013-09-16 |
Reductions show Max‑3SAT and planted‑clique become problems of maximizing a cubic form over unit vectors by encoding clauses or parity into an n×n×n tensor; a… | ||
| Theory and Big Data | Big Data Boot Camp 2013 | 2013-09-03 |
Argues that linear algebra + randomness actually gives provable wins: computing the best-fit k-subspace via SVD and projecting reduces dimension and often acts… | ||
| nothing matches | ||