AI Paper Atlas

Research map / Learning theory and algorithms / Sample complexity

Order Optimal One-Shot Distributed Learning

Arsalan Sharifnassab, Saber Salehkaleybar, S. Jamaloddin Golestani

NeurIPS 2019 · Learning theory and algorithms › Sample complexity · approach: Statistical-query lower bounds · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use MRE-C-log to communicate empirical minimizers and multiresolution gradient informationOne-shot distributed convex optimization · General convex optimizationexpected estimation error: O\big( m^{-{1}/{\max(d,2)}} n^{-1/2} \big)Improved
Use MRE-C-log instead of AVGMRidge regression, d=2 and n=1 · Ridge regressionaverage parameter errorImproved
Use MRE-C-log instead of AVGMLogistic regression, d=2 and n=1 · Logistic regressionaverage parameter errorImproved
Increase the number of machines while keeping n=1 and use MRE-C-logRidge regression and logistic regression · Ridge regression and logistic regressionaverage parameter errorImproved
Use a one-bit randomized estimatorOne-dimensional distributed optimization · One-dimensional, one-bit communicationroot mean squared error: O\left(\frac{1}{\sqrt{n}}+\frac{1}{\sqrt{m}}\right)Improved
Use the naive derivative-grid estimatorOne-dimensional optimization with n=1 · One-dimensional, n=1expected absolute estimation error: O\big((\log(m)/\sqrt[3]{m})^k\big)Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy