<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Communication Efficiency on Murtaza Rangwala</title><link>https://murtaza-hatim.com/tags/communication-efficiency/</link><description>Recent content in Communication Efficiency on Murtaza Rangwala</description><generator>Hugo -- 0.147.2</generator><language>en</language><lastBuildDate>Sat, 26 Sep 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://murtaza-hatim.com/tags/communication-efficiency/index.xml" rel="self" type="application/rss+xml"/><item><title>Screen Before You Fetch: Compressed Byzantine Screening for Decentralized Learning on the Edge-Cloud Continuum</title><link>https://murtaza-hatim.com/publications/sketchguard/</link><pubDate>Sat, 26 Sep 2026 00:00:00 +0000</pubDate><guid>https://murtaza-hatim.com/publications/sketchguard/</guid><description>SketchGuard makes Byzantine-robust decentralized federated learning communication-efficient by screening neighbors on compact sketches, with three rules that provably close the attacks sketching opens, matching the accuracy of the defenses it wraps while cutting bytes by up to 41% and emulated edge-cloud round time by 39%; under review at the Journal of Parallel and Distributed Computing (JPDC).</description></item></channel></rss>