<?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>Generalization on Tolga Dimlioglu</title><link>https://tolgadimli.github.io/tags/generalization/</link><description>Recent content in Generalization on Tolga Dimlioglu</description><generator>Hugo -- 0.147.2</generator><language>en</language><lastBuildDate>Tue, 15 Jul 2025 00:00:00 +0000</lastBuildDate><atom:link href="https://tolgadimli.github.io/tags/generalization/index.xml" rel="self" type="application/rss+xml"/><item><title>Communication-Efficient Distributed Training for Collaborative Flat Optima Recovery in Deep Learning</title><link>https://tolgadimli.github.io/papers/paper2/</link><pubDate>Tue, 15 Jul 2025 00:00:00 +0000</pubDate><guid>https://tolgadimli.github.io/papers/paper2/</guid><description>We propose a communication-efficient distributed training framework (DPPF) that introduces a pushing force to promote wide minima recovery, improving generalization without incurring significant computational or communication overhead.</description></item></channel></rss>