V
Veeranjaneyulu Sadhanala
Researcher at Carnegie Mellon University
Publications - 14
Citations - 269
Veeranjaneyulu Sadhanala is an academic researcher from Carnegie Mellon University. The author has contributed to research in topics: Laplacian smoothing & Minimax. The author has an hindex of 9, co-authored 12 publications receiving 226 citations. Previous affiliations of Veeranjaneyulu Sadhanala include Indian Institute of Technology Bombay.
Papers
More filters
Proceedings Article
Graph Sparsification Approaches for Laplacian Smoothing
TL;DR: This work considers fitting the statistical estimate using a sparsified surrogate graph G, which shares the vertices of G but has far fewer edges, and is thus more tractable to work with computationally.
Posted Content
Total Variation Classes Beyond 1d: Minimax Rates, and the Limitations of Linear Smoothers
TL;DR: In this article, the authors consider the problem of estimating a function defined over $n$ locations on a $d$-dimensional grid and derive the minimax optimal (squared) $\ell_2$ estimation error rate, parametrized by $n and $C_n.
Posted Content
Parallel and Distributed Block-Coordinate Frank-Wolfe Algorithms
TL;DR: A notable feature of Frank-Wolfe algorithms is that they do not depend on worst-case bounded delays, but only (mildly) on expected delays, making them robust to stragglers and faulty worker threads.
Journal ArticleDOI
Additive models with trend filtering
TL;DR: A new backfitting algorithm whose iterations can be run in parallel is described, which (as far as the authors know) is the first of its kind and derived fast error rates for additive trend filtering estimates.
Proceedings Article
Parallel and distributed block-coordinate frank-wolfe algorithms
TL;DR: In this article, the authors study parallel and distributed Frank-Wolfe algorithms on shared memory machines with mini-batching, and the latter in a delayed update framework, and show significant speedups over competing state-of-the-art (and synchronous) methods.