<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Gianluca Kosmella | PHINDER EIC Project</title><link>https://phinder-eic.github.io/author/gianluca-kosmella/</link><atom:link href="https://phinder-eic.github.io/author/gianluca-kosmella/index.xml" rel="self" type="application/rss+xml"/><description>Gianluca Kosmella</description><generator>Hugo Blox Builder (https://hugoblox.com)</generator><language>en-us</language><lastBuildDate>Mon, 23 Mar 2026 00:00:00 +0000</lastBuildDate><image><url>https://phinder-eic.github.io/media/icon_hu_526da948f480ce3f.png</url><title>Gianluca Kosmella</title><link>https://phinder-eic.github.io/author/gianluca-kosmella/</link></image><item><title>Robustness-oriented training for automated design of photonic tensor cores</title><link>https://phinder-eic.github.io/publication/kosmella-2026-photonic-tensor-cores/</link><pubDate>Mon, 23 Mar 2026 00:00:00 +0000</pubDate><guid>https://phinder-eic.github.io/publication/kosmella-2026-photonic-tensor-cores/</guid><description>&lt;h2 id="relevance-to-phinder"&gt;Relevance to PHINDER&lt;/h2&gt;
&lt;p&gt;Robustness-oriented training connects model optimisation with the physical characteristics of photonic tensor cores.&lt;/p&gt;
&lt;p&gt;This publication is background research for the project.&lt;/p&gt;</description></item></channel></rss>